Add win32 wheel

.gitignore

@@ -10,7 +10,7 @@
 
 # Gradle files
 .gradle/
-build/
+android/build/
 gradlew
 gradlew.bat
 gradle
@@ -27,12 +27,8 @@ wheelhouse
 __pycache__
 *.egg-info
 python/dist
-python/vosk/*.cc
-python/vosk/*.c
-python/vosk/*.h
-python/vosk/*.i
-python/vosk/vosk.py
-python/vosk/vosk_wrap.cpp
+python/build
+python/vosk/*.so
 python/test/db
 python/test/hyp
 python/test/model
@@ -48,12 +44,13 @@ java/model-spk/
 java/model/
 
 # CSharp
-csharp/gen
-csharp/*.exe
-csharp/*.c
-csharp/model/
-csharp/test.wav
-Vosk.dll
+*.dll
+*.so
+*.nupkg
+csharp/demo/model
+csharp/demo/test.wav
+csharp/demo/bin
+csharp/demo/obj
 
 # Node
 nodejs/vosk_wrap.cc
@@ -62,3 +59,10 @@ nodejs/example/test.wav
 nodejs/node_modules
 nodejs/package-lock.json
 nodejs/build
+
+# C
+c/test_vosk
+c/test_vosk_speaker
+c/oprofile_data
+c/model
+c/test.wav

android/CMakeLists.txt

@@ -32,7 +32,7 @@ set(API_SOURCES
 "${PROJECT_SOURCE_DIR}/../src/vosk_api.h"
 )
 
-set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -O3 -DFST_NO_DYNAMIC_LINKING")
+set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -std=c++17 -O3 -DFST_NO_DYNAMIC_LINKING")
 
 add_library( vosk_jni SHARED
              build/generated-src/cpp/vosk_wrap.cc
@@ -60,7 +60,7 @@ target_link_libraries( vosk_jni
              ${KALDI_ROOT}/src/base/kaldi-base.a
              ${LIB_ROOT}/lib/libfst.a
              ${LIB_ROOT}/lib/libfstngram.a
-             ${LIB_ROOT}/lib/libopenblas_${OPENBLAS_ARCH}-r0.3.7.a
+             ${LIB_ROOT}/lib/libopenblas.a
              ${LIB_ROOT}/lib/libclapack.a
              ${LIB_ROOT}/lib/liblapack.a
              ${LIB_ROOT}/lib/libblas.a

android/build-kaldi.sh

@@ -39,7 +39,7 @@ WORKDIR_X86_64=`pwd`/build/kaldi_x86_64
 WORKDIR_ARM32=`pwd`/build/kaldi_arm_32
 WORKDIR_ARM64=`pwd`/build/kaldi_arm_64
 PATH=$PATH:$ANDROID_NDK_HOME/toolchains/llvm/prebuilt/${OS_NAME}-x86_64/bin
-OPENFST_VERSION=1.6.7
+OPENFST_VERSION=1.8.0
 
 mkdir -p $WORKDIR_ARM64/local/lib $WORKDIR_ARM32/local/lib $WORKDIR_X86_64/local/lib $WORKDIR_X86/local/lib
 
@@ -106,7 +106,7 @@ esac
 
 # openblas first
 cd $WORKDIR
-git clone -b v0.3.7 --single-branch https://github.com/xianyi/OpenBLAS
+git clone -b v0.3.13 --single-branch https://github.com/xianyi/OpenBLAS
 make -C OpenBLAS TARGET=$BLAS_ARCH ONLY_CBLAS=1 AR=$AR CC=$CC HOSTCC=gcc ARM_SOFTFP_ABI=1 USE_THREAD=0 NUM_THREADS=1 -j4
 make -C OpenBLAS install PREFIX=$WORKDIR/local
 
@@ -130,7 +130,7 @@ if [ "`uname`" == "Darwin"  ]; then
 fi
 
 CXX=$CXX CXXFLAGS="$ARCHFLAGS -O3 -DFST_NO_DYNAMIC_LINKING" ./configure --use-cuda=no \
-    --mathlib=OPENBLAS --shared \
+    --mathlib=OPENBLAS_CLAPACK --shared \
     --android-incdir=${ANDROID_TOOLCHAIN_PATH}/sysroot/usr/include \
     --host=$HOST --openblas-root=${WORKDIR}/local \
     --fst-root=${WORKDIR}/local --fst-version=${OPENFST_VERSION}

android/build.gradle

@@ -14,7 +14,7 @@ plugins {
 }
 
 def archiveName = "vosk-android"
-def libVersion = "0.3.16"
+def libVersion = "0.3.17"
 
 allprojects {
   repositories {

android/model-en/build.gradle

@@ -14,7 +14,7 @@ plugins {
 }
 
 def archiveName = "vosk-model-en"
-def libVersion = "0.3.16"
+def libVersion = "0.3.17"
 
 allprojects {
   repositories {

android/model-en/src/main/assets/vosk-model-en/README

@@ -1,2 +1,7 @@
-US English model for android
+US English model for mobile Vosk applications
 
+Copyright 2020 Alpha Cephei Inc
+
+Accuracy: 10.38 (tedlium test) 9.85 (librispeech test-clean)
+Speed: 0.11xRT (desktop)
+Latency: 0.15s (right context)

c/Makefile

@@ -1,53 +1,16 @@
-KALDI_ROOT=$(HOME)/travis/kaldi
-
-VOSK_SOURCES= \
-    ../src/kaldi_recognizer.cc \
-    ../src/language_model.cc \
-    ../src/model.cc \
-    ../src/spk_model.cc \
-    ../src/vosk_api.cc
-
-CFLAGS=-g -O2 -DFST_NO_DYNAMIC_LINKING -I../src -I$(KALDI_ROOT)/src -I$(KALDI_ROOT)/tools/openfst/include
-LIBS= \
-	$(KALDI_ROOT)/src/online2/kaldi-online2.a \
-	$(KALDI_ROOT)/src/decoder/kaldi-decoder.a \
-	$(KALDI_ROOT)/src/ivector/kaldi-ivector.a \
-	$(KALDI_ROOT)/src/gmm/kaldi-gmm.a \
-	$(KALDI_ROOT)/src/nnet3/kaldi-nnet3.a \
-	$(KALDI_ROOT)/src/tree/kaldi-tree.a \
-	$(KALDI_ROOT)/src/feat/kaldi-feat.a \
-	$(KALDI_ROOT)/src/lat/kaldi-lat.a \
-	$(KALDI_ROOT)/src/lm/kaldi-lm.a \
-	$(KALDI_ROOT)/src/hmm/kaldi-hmm.a \
-	$(KALDI_ROOT)/src/transform/kaldi-transform.a \
-	$(KALDI_ROOT)/src/cudamatrix/kaldi-cudamatrix.a \
-	$(KALDI_ROOT)/src/matrix/kaldi-matrix.a \
-	$(KALDI_ROOT)/src/fstext/kaldi-fstext.a \
-	$(KALDI_ROOT)/src/util/kaldi-util.a \
-	$(KALDI_ROOT)/src/base/kaldi-base.a \
-	$(KALDI_ROOT)/tools/OpenBLAS/install/lib/libopenblas.a \
-	$(KALDI_ROOT)/tools/OpenBLAS/install/lib/liblapack.a \
-	$(KALDI_ROOT)/tools/OpenBLAS/install/lib/libblas.a \
-	$(KALDI_ROOT)/tools/OpenBLAS/install/lib/libf2c.a \
-	$(KALDI_ROOT)/tools/openfst/lib/libfst.a \
-	$(KALDI_ROOT)/tools/openfst/lib/libfstngram.a
+CFLAGS=-I../src
+LDFLAGS=-L../src -lvosk -ldl -lpthread -Wl,-rpath=../src
 
 all: test_vosk test_vosk_speaker
 
-test_vosk: test_vosk.o libvosk.a
-	g++ $^ -o $@ $(LIBS) -lpthread
+test_vosk: test_vosk.o
+	g++ $^ -o $@ $(LDFLAGS)
 
-test_vosk_speaker: test_vosk_speaker.o libvosk.a
-	g++ $^ -o $@ $(LIBS) -lpthread
-
-libvosk.a: $(VOSK_SOURCES:.cc=.o)
-	ar rcs $@ $^
+test_vosk_speaker: test_vosk_speaker.o
+	g++ $^ -o $@ $(LDFLAGS)
 
 %.o: %.c
 	g++ $(CFLAGS) -c -o $@ $<
 
-%.o: %.cc
-	g++ -std=c++17 $(CFLAGS) -c -o $@ $<
-
 clean:
 	rm -f *.o *.a test_vosk test_vosk_speaker

csharp/Makefile

@@ -1,68 +0,0 @@
-KALDI_ROOT ?= $(HOME)/kaldi
-CFLAGS :=  -std=c++17 -g -O2 -DPIC -fPIC -Wno-unused-function
-CPPFLAGS := -I$(KALDI_ROOT)/src -I$(KALDI_ROOT)/tools/openfst/include -I../src -DFST_NO_DYNAMIC_LINKING
-
-KALDI_LIBS = \
-             ${KALDI_ROOT}/src/online2/kaldi-online2.a \
-             ${KALDI_ROOT}/src/decoder/kaldi-decoder.a \
-             ${KALDI_ROOT}/src/ivector/kaldi-ivector.a \
-             ${KALDI_ROOT}/src/gmm/kaldi-gmm.a \
-             ${KALDI_ROOT}/src/nnet3/kaldi-nnet3.a \
-             ${KALDI_ROOT}/src/tree/kaldi-tree.a \
-             ${KALDI_ROOT}/src/feat/kaldi-feat.a \
-             ${KALDI_ROOT}/src/lat/kaldi-lat.a \
-             ${KALDI_ROOT}/src/lm/kaldi-lm.a \
-             ${KALDI_ROOT}/src/hmm/kaldi-hmm.a \
-             ${KALDI_ROOT}/src/transform/kaldi-transform.a \
-             ${KALDI_ROOT}/src/cudamatrix/kaldi-cudamatrix.a \
-             ${KALDI_ROOT}/src/matrix/kaldi-matrix.a \
-             ${KALDI_ROOT}/src/fstext/kaldi-fstext.a \
-             ${KALDI_ROOT}/src/util/kaldi-util.a \
-             ${KALDI_ROOT}/src/base/kaldi-base.a \
-             ${KALDI_ROOT}/tools/openfst/lib/libfst.a \
-             ${KALDI_ROOT}/tools/openfst/lib/libfstngram.a
-
-# On Linux
-MATH_LIBS = \
-    ${KALDI_ROOT}/tools/OpenBLAS/install/lib/libopenblas.a \
-    ${KALDI_ROOT}/tools/OpenBLAS/install/lib/liblapack.a \
-    ${KALDI_ROOT}/tools/OpenBLAS/install/lib/libblas.a \
-    ${KALDI_ROOT}/tools/OpenBLAS/install/lib/libf2c.a
-
-# On OSX
-# MATH_LIBS = -framework Accelerate
-
-all: test.exe
-
-test.exe: libkaldiwrap.so test.cs
-	mcs -out:Vosk.dll -target:library gen/*.cs
-	mcs test.cs -r:Vosk.dll -out:test.exe
-
-VOSK_SOURCES = \
-	vosk_wrap.c \
-	../src/kaldi_recognizer.cc \
-	../src/language_model.cc \
-	../src/model.cc \
-	../src/spk_model.cc \
-	../src/vosk_api.cc
-
-VOSK_HEADERS = \
-	../src/kaldi_recognizer.h \
-	../src/language_model.h \
-	../src/model.h \
-	../src/vosk_api.h \
-	../src/spk_model.h
-
-libkaldiwrap.so: $(VOSK_SOURCES) $(VOSK_HEADERS)
-	$(CXX) -fpermissive $(CFLAGS) $(CPPFLAGS) -shared -o $@ $(VOSK_SOURCES) $(KALDI_LIBS) $(MATH_LIBS)
-
-vosk_wrap.c: ../src/vosk.i
-	mkdir -p gen
-	swig -csharp -DSWIG_CSHARP_NO_EXCEPTION_HELPER -dllimport "libkaldiwrap" \
-	    -namespace "Kaldi" -outdir gen -o vosk_wrap.c ../src/vosk.i
-
-run: test.exe
-	mono test.exe
-
-clean:
-	$(RM) *.so vosk_wrap.c *.o gen/*.cs test.exe *.dll

csharp/README.md

@@ -0,0 +1,12 @@
+This is a nuget-based wrapper for libvosk library
+
+See demo folder for example how to use the library. You can simply run
+"dotnet run" to run the demo. Make sure you unpacked the model and the
+test file.
+
+See the nuget folder for the sources of the wrapper. Run build.sh to
+build nuget package.
+
+Note we only support win64 and linux64 for now. No support for win32
+since it is a little bit painful to load the libraries depending on
+architecture. In theory we can add OSX some time or even Android.

csharp/demo/VoskDemo.cs

@@ -1,15 +1,15 @@
 using System;
 using System.IO;
-using Kaldi;
+using Vosk;
 
 public class Test
 {
    public static void Main()
    {
 
-        Vosk.SetLogLevel(0);
+        Vosk.Vosk.SetLogLevel(0);
         Model model = new Model("model");
-        KaldiRecognizer rec = new KaldiRecognizer(model, 16000.0f);
+        VoskRecognizer rec = new VoskRecognizer(model, 16000.0f);
 
         using(Stream source = File.OpenRead("test.wav")) {
             byte[] buffer = new byte[4096];
@@ -24,7 +24,7 @@ public class Test
         }
         Console.WriteLine(rec.FinalResult());
 
-        rec = new KaldiRecognizer(model, 16000.0f);
+        rec = new VoskRecognizer(model, 16000.0f);
 
         using(Stream source = File.OpenRead("test.wav")) {
             byte[] buffer = new byte[4096];

csharp/demo/VoskDemo.csproj

@@ -0,0 +1,13 @@
+<Project Sdk="Microsoft.NET.Sdk">
+
+  <PropertyGroup>
+    <OutputType>Exe</OutputType>
+    <TargetFramework>net5.0</TargetFramework>
+    <RootNamespace>VoskDemo</RootNamespace>
+  </PropertyGroup>
+
+  <ItemGroup>
+    <PackageReference Include="Vosk" Version="0.3.18" />
+  </ItemGroup>
+
+</Project>

csharp/nuget/Vosk.nuspec

@@ -2,7 +2,7 @@
 <package>
   <metadata>
     <id>Vosk</id>
-    <version>0.3.16</version>
+    <version>0.3.18</version>
     <authors>Alpha Cephei Inc</authors>
     <owners>Alpha Cephei Inc</owners>
     <license type="expression">Apache-2.0</license>
@@ -24,4 +24,7 @@ Vosk scales from small devices like Raspberry Pi or Android smartphone to big cl
       <group targetFramework=".NETStandard2.0"/>
     </dependencies>
   </metadata>
+  <files>
+    <file src="**" exclude="src/*.cs;build.sh;**/.keep-me;*.nupkg" />
+  </files>
 </package>

csharp/nuget/build.sh

@@ -0,0 +1,2 @@
+mcs -out:lib/netstandard2.0/Vosk.dll -target:library src/*.cs
+nuget pack

csharp/nuget/build/Vosk.targets

@@ -0,0 +1,10 @@
+<Project xmlns="http://schemas.microsoft.com/developer/msbuild/2003">
+  <ItemGroup>
+    <NativeLibs Include="$(MSBuildThisFileDirectory)\lib\linux-x64\*.so" Condition="'$([MSBuild]::IsOsPlatform(Linux))'" />
+    <NativeLibs Include="$(MSBuildThisFileDirectory)\lib\win-x64\*.dll" Condition="'$([MSBuild]::IsOsPlatform(Windows))'" />
+    <None Include="@(NativeLibs)">
+      <Link>%(FileName)%(Extension)</Link>
+      <CopyToOutputDirectory>PreserveNewest</CopyToOutputDirectory>
+    </None>
+  </ItemGroup>
+</Project>

csharp/nuget/build/netstandard2.0/Vosk.targets

@@ -1,9 +0,0 @@
-<Project xmlns="http://schemas.microsoft.com/developer/msbuild/2003">
-	<ItemGroup>
-		<Content Include="$(MSBuildThisFileDirectory)lib/**/**">
-			<Link>%(RecursiveDir)%(Filename)%(Extension)</Link>
-			<CopyToOutputDirectory>PreserveNewest</CopyToOutputDirectory>
-			<Visible>false</Visible>
-		</Content>
-	</ItemGroup>
-</Project>

csharp/nuget/src/Model.cs

@@ -0,0 +1,41 @@
+namespace Vosk {
+
+public class Model : global::System.IDisposable {
+  private global::System.Runtime.InteropServices.HandleRef handle;
+
+  internal Model(global::System.IntPtr cPtr) {
+    handle = new global::System.Runtime.InteropServices.HandleRef(this, cPtr);
+  }
+
+  internal static global::System.Runtime.InteropServices.HandleRef getCPtr(Model obj) {
+    return (obj == null) ? new global::System.Runtime.InteropServices.HandleRef(null, global::System.IntPtr.Zero) : obj.handle;
+  }
+
+  ~Model() {
+    Dispose(false);
+  }
+
+  public void Dispose() {
+    Dispose(true);
+    global::System.GC.SuppressFinalize(this);
+  }
+
+  protected virtual void Dispose(bool disposing) {
+    lock(this) {
+      if (handle.Handle != global::System.IntPtr.Zero) {
+        VoskPINVOKE.delete_Model(handle);
+        handle = new global::System.Runtime.InteropServices.HandleRef(null, global::System.IntPtr.Zero);
+      }
+    }
+  }
+
+  public Model(string model_path) : this(VoskPINVOKE.new_Model(model_path)) {
+  }
+
+  public int vosk_model_find_word(string word) {
+    return VoskPINVOKE.Model_vosk_model_find_word(handle, word);
+  }
+
+}
+
+}

csharp/nuget/src/SpkModel.cs

@@ -0,0 +1,37 @@
+namespace Vosk {
+
+public class SpkModel : global::System.IDisposable {
+  private global::System.Runtime.InteropServices.HandleRef handle;
+
+  internal SpkModel(global::System.IntPtr cPtr) {
+    handle = new global::System.Runtime.InteropServices.HandleRef(this, cPtr);
+  }
+
+  internal static global::System.Runtime.InteropServices.HandleRef getCPtr(SpkModel obj) {
+    return (obj == null) ? new global::System.Runtime.InteropServices.HandleRef(null, global::System.IntPtr.Zero) : obj.handle;
+  }
+
+  ~SpkModel() {
+    Dispose(false);
+  }
+
+  public void Dispose() {
+    Dispose(true);
+    global::System.GC.SuppressFinalize(this);
+  }
+
+  protected virtual void Dispose(bool disposing) {
+    lock(this) {
+      if (handle.Handle != global::System.IntPtr.Zero) {
+        VoskPINVOKE.delete_SpkModel(handle);
+        handle = new global::System.Runtime.InteropServices.HandleRef(null, global::System.IntPtr.Zero);
+      }
+    }
+  }
+
+  public SpkModel(string model_path) : this(VoskPINVOKE.new_SpkModel(model_path)) {
+  }
+
+}
+
+}

csharp/nuget/src/Vosk.cs

@@ -0,0 +1,10 @@
+namespace Vosk {
+
+public class Vosk {
+  public static void SetLogLevel(int level) {
+    VoskPINVOKE.SetLogLevel(level);
+  }
+
+}
+
+}

csharp/nuget/src/VoskPINVOKE.cs

@@ -0,0 +1,57 @@
+namespace Vosk {
+
+class VoskPINVOKE {
+
+  static VoskPINVOKE() {
+  }
+
+  [global::System.Runtime.InteropServices.DllImport("libvosk", EntryPoint="vosk_model_new")]
+  public static extern global::System.IntPtr new_Model(string jarg1);
+
+  [global::System.Runtime.InteropServices.DllImport("libvosk", EntryPoint="vosk_model_free")]
+  public static extern void delete_Model(global::System.Runtime.InteropServices.HandleRef jarg1);
+
+  [global::System.Runtime.InteropServices.DllImport("libvosk", EntryPoint="vosk_model_find_word")]
+  public static extern int Model_vosk_model_find_word(global::System.Runtime.InteropServices.HandleRef jarg1, string jarg2);
+
+  [global::System.Runtime.InteropServices.DllImport("libvosk", EntryPoint="vosk_spk_model_new")]
+  public static extern global::System.IntPtr new_SpkModel(string jarg1);
+
+  [global::System.Runtime.InteropServices.DllImport("libvosk", EntryPoint="vosk_spk_model_free")]
+  public static extern void delete_SpkModel(global::System.Runtime.InteropServices.HandleRef jarg1);
+
+  [global::System.Runtime.InteropServices.DllImport("libvosk", EntryPoint="vosk_recognizer_new")]
+  public static extern global::System.IntPtr new_VoskRecognizer(global::System.Runtime.InteropServices.HandleRef jarg1, float jarg2);
+
+  [global::System.Runtime.InteropServices.DllImport("libvosk", EntryPoint="vosk_recognizer_new_spk")]
+  public static extern global::System.IntPtr new_VoskRecognizerSpk(global::System.Runtime.InteropServices.HandleRef jarg1, global::System.Runtime.InteropServices.HandleRef jarg2, float jarg3);
+
+  [global::System.Runtime.InteropServices.DllImport("libvosk", EntryPoint="vosk_recognizer_new_grm")]
+  public static extern global::System.IntPtr new_VoskRecognizerGrm(global::System.Runtime.InteropServices.HandleRef jarg1, float jarg2, string jarg3);
+
+  [global::System.Runtime.InteropServices.DllImport("libvosk", EntryPoint="vosk_recognizer_free")]
+  public static extern void delete_VoskRecognizer(global::System.Runtime.InteropServices.HandleRef jarg1);
+
+  [global::System.Runtime.InteropServices.DllImport("libvosk", EntryPoint="vosk_recognizer_accept_waveform")]
+  public static extern bool VoskRecognizer_AcceptWaveform(global::System.Runtime.InteropServices.HandleRef jarg1, [global::System.Runtime.InteropServices.In, global::System.Runtime.InteropServices.MarshalAs(global::System.Runtime.InteropServices.UnmanagedType.LPArray)]byte[] jarg2, int jarg3);
+
+  [global::System.Runtime.InteropServices.DllImport("libvosk", EntryPoint="vosk_recognizer_accept_waveform_s")]
+  public static extern bool VoskRecognizer_AcceptWaveformShort(global::System.Runtime.InteropServices.HandleRef jarg1, [global::System.Runtime.InteropServices.In, global::System.Runtime.InteropServices.MarshalAs(global::System.Runtime.InteropServices.UnmanagedType.LPArray)]short[] jarg2, int jarg3);
+
+  [global::System.Runtime.InteropServices.DllImport("libvosk", EntryPoint="vosk_recognizer_accept_waveform_f")]
+  public static extern bool VoskRecognizer_AcceptWaveformFloat(global::System.Runtime.InteropServices.HandleRef jarg1, [global::System.Runtime.InteropServices.In, global::System.Runtime.InteropServices.MarshalAs(global::System.Runtime.InteropServices.UnmanagedType.LPArray)]float[] jarg2, int jarg3);
+
+  [global::System.Runtime.InteropServices.DllImport("libvosk", EntryPoint="vosk_recognizer_result")]
+  public static extern global::System.IntPtr VoskRecognizer_Result(global::System.Runtime.InteropServices.HandleRef jarg1);
+
+  [global::System.Runtime.InteropServices.DllImport("libvosk", EntryPoint="vosk_recognizer_partial_result")]
+  public static extern global::System.IntPtr VoskRecognizer_PartialResult(global::System.Runtime.InteropServices.HandleRef jarg1);
+
+  [global::System.Runtime.InteropServices.DllImport("libvosk", EntryPoint="vosk_recognizer_final_result")]
+  public static extern global::System.IntPtr VoskRecognizer_FinalResult(global::System.Runtime.InteropServices.HandleRef jarg1);
+
+  [global::System.Runtime.InteropServices.DllImport("libvosk", EntryPoint="vosk_set_log_level")]
+  public static extern void SetLogLevel(int jarg1);
+}
+
+}

csharp/nuget/src/VoskRecognizer.cs

@@ -0,0 +1,67 @@
+namespace Vosk {
+
+public class VoskRecognizer : global::System.IDisposable {
+  private global::System.Runtime.InteropServices.HandleRef handle;
+
+  internal VoskRecognizer(global::System.IntPtr cPtr) {
+    handle = new global::System.Runtime.InteropServices.HandleRef(this, cPtr);
+  }
+
+  internal static global::System.Runtime.InteropServices.HandleRef getCPtr(VoskRecognizer obj) {
+    return (obj == null) ? new global::System.Runtime.InteropServices.HandleRef(null, global::System.IntPtr.Zero) : obj.handle;
+  }
+
+  ~VoskRecognizer() {
+    Dispose(false);
+  }
+
+  public void Dispose() {
+    Dispose(true);
+    global::System.GC.SuppressFinalize(this);
+  }
+
+  protected virtual void Dispose(bool disposing) {
+    lock(this) {
+      if (handle.Handle != global::System.IntPtr.Zero) {
+        VoskPINVOKE.delete_VoskRecognizer(handle);
+        handle = new global::System.Runtime.InteropServices.HandleRef(null, global::System.IntPtr.Zero);
+      }
+    }
+  }
+
+  public VoskRecognizer(Model model, float sample_rate) : this(VoskPINVOKE.new_VoskRecognizer(Model.getCPtr(model), sample_rate)) {
+  }
+
+  public VoskRecognizer(Model model, SpkModel spk_model, float sample_rate) : this(VoskPINVOKE.new_VoskRecognizerSpk(Model.getCPtr(model), SpkModel.getCPtr(spk_model), sample_rate)) {
+  }
+
+  public VoskRecognizer(Model model, float sample_rate, string grammar) : this(VoskPINVOKE.new_VoskRecognizerGrm(Model.getCPtr(model), sample_rate, grammar)) {
+  }
+
+  public bool AcceptWaveform(byte[] data, int len) {
+    return VoskPINVOKE.VoskRecognizer_AcceptWaveform(handle, data, len);
+  }
+
+  public bool AcceptWaveform(short[] sdata, int len) {
+    return VoskPINVOKE.VoskRecognizer_AcceptWaveformShort(handle, sdata, len);
+  }
+
+  public bool AcceptWaveform(float[] fdata, int len) {
+    return VoskPINVOKE.VoskRecognizer_AcceptWaveformFloat(handle, fdata, len);
+  }
+
+  public string Result() {
+    return global::System.Runtime.InteropServices.Marshal.PtrToStringUTF8(VoskPINVOKE.VoskRecognizer_Result(handle));
+  }
+
+  public string PartialResult() {
+    return global::System.Runtime.InteropServices.Marshal.PtrToStringUTF8(VoskPINVOKE.VoskRecognizer_PartialResult(handle));
+  }
+
+  public string FinalResult() {
+    return global::System.Runtime.InteropServices.Marshal.PtrToStringUTF8(VoskPINVOKE.VoskRecognizer_FinalResult(handle));
+  }
+
+}
+
+}

ios/VoskApiTest.xcodeproj/project.pbxproj

@@ -17,7 +17,8 @@
 		92375244240C6DAF00DD6076 /* Accelerate.framework in Frameworks */ = {isa = PBXBuildFile; fileRef = 92375243240C6DAF00DD6076 /* Accelerate.framework */; };
 		92375246240C6DC900DD6076 /* libstdc++.tbd in Frameworks */ = {isa = PBXBuildFile; fileRef = 92375245240C6DC900DD6076 /* libstdc++.tbd */; };
 		92375274240C6F1E00DD6076 /* 10001-90210-01803.wav in Resources */ = {isa = PBXBuildFile; fileRef = 92375256240C6E3D00DD6076 /* 10001-90210-01803.wav */; };
-		92D86BD5253F823F0040D53F /* vosk-model-small-en-us-0.4 in Resources */ = {isa = PBXBuildFile; fileRef = 92D86BD3253F823E0040D53F /* vosk-model-small-en-us-0.4 */; };
+		92BACED125BE125A00B5CC93 /* vosk-model-small-en-us-0.15 in Resources */ = {isa = PBXBuildFile; fileRef = 928CC50C25BE124400490481 /* vosk-model-small-en-us-0.15 */; };
+		92D6B8D325BDFEAC007FF08D /* VoskModel.swift in Sources */ = {isa = PBXBuildFile; fileRef = 92D6B8D225BDFEAC007FF08D /* VoskModel.swift */; };
 		92D86BD6253F823F0040D53F /* vosk-model-spk-0.4 in Resources */ = {isa = PBXBuildFile; fileRef = 92D86BD4253F823F0040D53F /* vosk-model-spk-0.4 */; };
 /* End PBXBuildFile section */
 
@@ -34,9 +35,10 @@
 		92375243240C6DAF00DD6076 /* Accelerate.framework */ = {isa = PBXFileReference; lastKnownFileType = wrapper.framework; name = Accelerate.framework; path = System/Library/Frameworks/Accelerate.framework; sourceTree = SDKROOT; };
 		92375245240C6DC900DD6076 /* libstdc++.tbd */ = {isa = PBXFileReference; lastKnownFileType = "sourcecode.text-based-dylib-definition"; name = "libstdc++.tbd"; path = "usr/lib/libstdc++.tbd"; sourceTree = SDKROOT; };
 		92375256240C6E3D00DD6076 /* 10001-90210-01803.wav */ = {isa = PBXFileReference; lastKnownFileType = audio.wav; path = "10001-90210-01803.wav"; sourceTree = "<group>"; };
+		928CC50C25BE124400490481 /* vosk-model-small-en-us-0.15 */ = {isa = PBXFileReference; lastKnownFileType = folder; name = "vosk-model-small-en-us-0.15"; path = "/Users/shmyrev/Documents/IOS/VoskApiTest/VoskApiTest/Vosk/vosk-model-small-en-us-0.15"; sourceTree = "<absolute>"; };
 		92AA22AD244CDD1200DA464B /* vosk_api.h */ = {isa = PBXFileReference; fileEncoding = 4; lastKnownFileType = sourcecode.c.h; path = vosk_api.h; sourceTree = "<group>"; };
 		92AA22AE244CDD5200DA464B /* bridging.h */ = {isa = PBXFileReference; fileEncoding = 4; lastKnownFileType = sourcecode.c.h; path = bridging.h; sourceTree = "<group>"; };
-		92D86BD3253F823E0040D53F /* vosk-model-small-en-us-0.4 */ = {isa = PBXFileReference; lastKnownFileType = folder; path = "vosk-model-small-en-us-0.4"; sourceTree = "<group>"; };
+		92D6B8D225BDFEAC007FF08D /* VoskModel.swift */ = {isa = PBXFileReference; lastKnownFileType = sourcecode.swift; path = VoskModel.swift; sourceTree = "<group>"; };
 		92D86BD4253F823F0040D53F /* vosk-model-spk-0.4 */ = {isa = PBXFileReference; lastKnownFileType = folder; path = "vosk-model-spk-0.4"; sourceTree = "<group>"; };
 /* End PBXFileReference section */
 
@@ -83,6 +85,7 @@
 				9237522A240C550B00DD6076 /* LaunchScreen.storyboard */,
 				9237522D240C550B00DD6076 /* Info.plist */,
 				92375233240C558900DD6076 /* Vosk.swift */,
+				92D6B8D225BDFEAC007FF08D /* VoskModel.swift */,
 			);
 			path = VoskApiTest;
 			sourceTree = "<group>";
@@ -90,7 +93,7 @@
 		92375239240C642000DD6076 /* Vosk */ = {
 			isa = PBXGroup;
 			children = (
-				92D86BD3253F823E0040D53F /* vosk-model-small-en-us-0.4 */,
+				928CC50C25BE124400490481 /* vosk-model-small-en-us-0.15 */,
 				92D86BD4253F823F0040D53F /* vosk-model-spk-0.4 */,
 				92375256240C6E3D00DD6076 /* 10001-90210-01803.wav */,
 				92AA22AD244CDD1200DA464B /* vosk_api.h */,
@@ -169,12 +172,12 @@
 			isa = PBXResourcesBuildPhase;
 			buildActionMask = 2147483647;
 			files = (
+				92BACED125BE125A00B5CC93 /* vosk-model-small-en-us-0.15 in Resources */,
 				92375274240C6F1E00DD6076 /* 10001-90210-01803.wav in Resources */,
 				9237522C240C550B00DD6076 /* LaunchScreen.storyboard in Resources */,
 				92375229240C550B00DD6076 /* Assets.xcassets in Resources */,
 				92D86BD6253F823F0040D53F /* vosk-model-spk-0.4 in Resources */,
 				92375227240C550B00DD6076 /* Main.storyboard in Resources */,
-				92D86BD5253F823F0040D53F /* vosk-model-small-en-us-0.4 in Resources */,
 			);
 			runOnlyForDeploymentPostprocessing = 0;
 		};
@@ -187,6 +190,7 @@
 			files = (
 				92375224240C550B00DD6076 /* ViewController.swift in Sources */,
 				92375222240C550B00DD6076 /* AppDelegate.swift in Sources */,
+				92D6B8D325BDFEAC007FF08D /* VoskModel.swift in Sources */,
 				92375234240C558900DD6076 /* Vosk.swift in Sources */,
 			);
 			runOnlyForDeploymentPostprocessing = 0;

ios/VoskApiTest.xcodeproj/project.xcworkspace/contents.xcworkspacedata

@@ -2,10 +2,7 @@
 <Workspace
    version = "1.0">
    <FileRef
-      location = "group:../VoskApiTest/Vosk/vosk-model-small-en-us-0.4">
-   </FileRef>
-   <FileRef
-      location = "group:../VoskApiTest/Vosk/vosk-model-spk-0.4">
+      location = "group:/Users/shmyrev/Documents/IOS/VoskApiTest/VoskApiTest/Vosk/vosk-model-small-en-us-0.15">
    </FileRef>
    <FileRef
       location = "self:">

ios/VoskApiTest/Base.lproj/Main.storyboard

@@ -1,6 +1,6 @@
 <?xml version="1.0" encoding="UTF-8"?>
-<document type="com.apple.InterfaceBuilder3.CocoaTouch.Storyboard.XIB" version="3.0" toolsVersion="13771" targetRuntime="iOS.CocoaTouch" propertyAccessControl="none" useAutolayout="YES" useTraitCollections="YES" colorMatched="YES" initialViewController="BYZ-38-t0r">
-    <device id="retina4_7" orientation="portrait">
+<document type="com.apple.InterfaceBuilder3.CocoaTouch.Storyboard.XIB" version="3.0" toolsVersion="13771" targetRuntime="iOS.CocoaTouch" propertyAccessControl="none" useAutolayout="YES" useTraitCollections="YES" colorMatched="YES" initialViewController="bdW-KL-Y8Z">
+    <device id="retina5_5" orientation="portrait">
         <adaptation id="fullscreen"/>
     </device>
     <dependencies>
@@ -10,23 +10,52 @@
     </dependencies>
     <scenes>
         <!--View Controller-->
-        <scene sceneID="tne-QT-ifu">
+        <scene sceneID="nEc-89-Iqu">
             <objects>
-                <viewController id="BYZ-38-t0r" customClass="ViewController" customModule="VoskApiTest" customModuleProvider="target" sceneMemberID="viewController">
-                    <textView key="view" clipsSubviews="YES" multipleTouchEnabled="YES" contentMode="scaleToFill" editable="NO" textAlignment="natural" id="CtX-mx-X98">
-                        <rect key="frame" x="0.0" y="0.0" width="375" height="667"/>
-                        <autoresizingMask key="autoresizingMask" flexibleMaxX="YES" flexibleMaxY="YES"/>
+                <viewController id="bdW-KL-Y8Z" customClass="ViewController" customModule="VoskApiTest" customModuleProvider="target" sceneMemberID="viewController">
+                    <layoutGuides>
+                        <viewControllerLayoutGuide type="top" id="Hyr-Dz-4mU"/>
+                        <viewControllerLayoutGuide type="bottom" id="w4A-5X-uBu"/>
+                    </layoutGuides>
+                    <view key="view" contentMode="scaleToFill" id="m5v-US-bvR">
+                        <rect key="frame" x="0.0" y="0.0" width="414" height="736"/>
+                        <autoresizingMask key="autoresizingMask" widthSizable="YES" heightSizable="YES"/>
+                        <subviews>
+                            <button opaque="NO" contentMode="scaleToFill" fixedFrame="YES" contentHorizontalAlignment="center" contentVerticalAlignment="center" buttonType="roundedRect" lineBreakMode="middleTruncation" translatesAutoresizingMaskIntoConstraints="NO" id="IaT-no-U3i" userLabel="Microphone">
+                                <rect key="frame" x="124" y="34" width="157" height="30"/>
+                                <autoresizingMask key="autoresizingMask" flexibleMaxX="YES" flexibleMaxY="YES"/>
+                                <state key="normal" title="Recognize Microphone"/>
+                                <connections>
+                                    <action selector="runRecognizeMicrohpone:" destination="bdW-KL-Y8Z" eventType="touchUpInside" id="hGB-lz-N2B"/>
+                                </connections>
+                            </button>
+                            <button opaque="NO" contentMode="scaleToFill" fixedFrame="YES" contentHorizontalAlignment="center" contentVerticalAlignment="center" buttonType="roundedRect" lineBreakMode="middleTruncation" translatesAutoresizingMaskIntoConstraints="NO" id="GC5-nT-FQR" userLabel="File">
+                                <rect key="frame" x="90" y="84" width="221" height="41"/>
+                                <autoresizingMask key="autoresizingMask" flexibleMaxX="YES" flexibleMaxY="YES"/>
+                                <state key="normal" title="Recognize File"/>
+                                <connections>
+                                    <action selector="runRecognizeFile:" destination="bdW-KL-Y8Z" eventType="touchUpInside" id="xp5-Yi-rnN"/>
+                                </connections>
+                            </button>
+                            <textView clipsSubviews="YES" multipleTouchEnabled="YES" contentMode="scaleToFill" fixedFrame="YES" text="Results here" textAlignment="natural" translatesAutoresizingMaskIntoConstraints="NO" id="w4X-cu-USq">
+                                <rect key="frame" x="11" y="112" width="383" height="569"/>
+                                <autoresizingMask key="autoresizingMask" flexibleMaxX="YES" flexibleMaxY="YES"/>
+                                <color key="backgroundColor" white="1" alpha="1" colorSpace="calibratedWhite"/>
+                                <fontDescription key="fontDescription" type="system" pointSize="14"/>
+                                <textInputTraits key="textInputTraits" autocapitalizationType="sentences"/>
+                            </textView>
+                        </subviews>
                         <color key="backgroundColor" white="1" alpha="1" colorSpace="calibratedWhite"/>
-                        <fontDescription key="fontDescription" type="system" pointSize="14"/>
-                        <textInputTraits key="textInputTraits" autocapitalizationType="sentences"/>
-                    </textView>
+                    </view>
                     <connections>
-                        <outlet property="mainText" destination="CtX-mx-X98" id="oJy-5J-NKp"/>
+                        <outlet property="mainText" destination="w4X-cu-USq" id="rZS-nz-Wql"/>
+                        <outlet property="recognizeFile" destination="GC5-nT-FQR" id="dRe-tc-IA0"/>
+                        <outlet property="recognizeMicrophone" destination="IaT-no-U3i" id="IuM-aa-pAP"/>
                     </connections>
                 </viewController>
-                <placeholder placeholderIdentifier="IBFirstResponder" id="dkx-z0-nzr" sceneMemberID="firstResponder"/>
+                <placeholder placeholderIdentifier="IBFirstResponder" id="nWA-4D-pA6" userLabel="First Responder" sceneMemberID="firstResponder"/>
             </objects>
-            <point key="canvasLocation" x="32.799999999999997" y="32.833583208395808"/>
+            <point key="canvasLocation" x="-17.39130434782609" y="-285.32608695652175"/>
         </scene>
     </scenes>
 </document>

ios/VoskApiTest/ViewController.swift

@@ -3,30 +3,113 @@
 //  VoskApiTest
 //
 //  Created by Niсkolay Shmyrev on 01.03.20.
-//  Copyright © 2020 Alpha Cephei. All rights reserved.
+//  Copyright © 2020-2021 Alpha Cephei. All rights reserved.
 //
 
 import UIKit
+import AVFoundation
+
+enum WorkMode {
+    case stopped
+    case microphone
+    case file
+}
 
 class ViewController: UIViewController {
-
-    @IBOutlet var mainText: UITextView!
-
-    override func viewDidLoad() {
-        super.viewDidLoad()
     
-        DispatchQueue.global(qos: .userInitiated).async {
-            DispatchQueue.main.async {
-                 self.mainText.text =  "Processing file..."
+    var mode: WorkMode!
+    
+    @IBOutlet weak var recognizeFile: UIButton!
+    @IBOutlet weak var mainText: UITextView!
+    @IBOutlet weak var recognizeMicrophone: UIButton!
+    
+    var audioEngine : AVAudioEngine!
+    var processingQueue: DispatchQueue!
+    var model : VoskModel!
+    
+    func setMode(mode: WorkMode) {
+        switch mode {
+        case .stopped:
+            self.recognizeFile.isEnabled = true
+            self.recognizeMicrophone.isEnabled = true
+            self.recognizeMicrophone.setTitle("Recognize Microphone",for: .normal)
+        case .microphone:
+            self.recognizeFile.isEnabled = false
+            self.recognizeMicrophone.isEnabled = true
+            self.recognizeMicrophone.setTitle("Stop Microphone",for: .normal)
+            self.mainText.text = ""
+        case .file:
+            self.recognizeFile.isEnabled = false
+            self.recognizeMicrophone.isEnabled = false
+            self.mainText.text =  "Processing file..."
+        }
+        self.mode = mode
+    }
+    
+    func startAudioEngine() {
+        do {
+            
+            // Create a new audio engine.
+            audioEngine = AVAudioEngine()
+            
+            let inputNode = audioEngine.inputNode
+            let formatInput = inputNode.inputFormat(forBus: 0)
+            let formatPcm = AVAudioFormat.init(commonFormat: AVAudioCommonFormat.pcmFormatInt16, sampleRate: formatInput.sampleRate, channels: 1, interleaved: true)
+            
+            let recognizer = Vosk(model: model, sampleRate: Float(formatInput.sampleRate))
+            
+            inputNode.installTap(onBus: 0,
+                                 bufferSize: UInt32(formatInput.sampleRate / 10),
+                                 format: formatPcm) { buffer, time in
+                                    self.processingQueue.async {
+                                        let res = recognizer.recognizeData(buffer: buffer)
+                                        DispatchQueue.main.async {
+                                            self.mainText.text = res + "\n" + self.mainText.text
+                                        }
+                                    }
             }
-            let vosk = Vosk()
-            let res = vosk.recognizeFile()
+            
+            // Start the stream of audio data.
+            audioEngine.prepare()
+            try audioEngine.start()
+        } catch {
+            print("Unable to start AVAudioEngine: \(error.localizedDescription)")
+        }
+    }
+    
+    func stopAudioEngine() {
+        audioEngine.stop()
+    }
+    
+    @IBAction func runRecognizeMicrohpone(_ sender: Any) {
+        if (mode == .stopped) {
+            setMode(mode: .microphone)
+            startAudioEngine()
+        } else {
+            stopAudioEngine()
+            setMode(mode: .stopped)
+        }
+    }
+    
+    @IBAction func runRecognizeFile(_ sender: Any) {
+        setMode(mode: .file)
+        processingQueue.async {
+            let recognizer = Vosk(model: self.model, sampleRate: 16000.0)
+            let res = recognizer.recognizeFile()
             DispatchQueue.main.async {
                 self.mainText.text = res
+                self.setMode(mode: .stopped)
             }
         }
     }
-
+    
+    override func viewDidLoad() {
+        super.viewDidLoad()
+        setMode(mode: .stopped)
+        processingQueue = DispatchQueue(label: "recognizerQueue")
+        model = VoskModel()
+    }
+    
     override func didReceiveMemoryWarning() {
         super.didReceiveMemoryWarning()
     }

ios/VoskApiTest/Vosk.swift

@@ -3,40 +3,52 @@
 //  VoskApiTest
 //
 //  Created by Niсkolay Shmyrev on 01.03.20.
-//  Copyright © 2020 Alpha Cephei. All rights reserved.
+//  Copyright © 2020-2021 Alpha Cephei. All rights reserved.
 //
 
 import Foundation
+import AVFoundation
 
 public final class Vosk {
     
+    var recognizer : OpaquePointer!
+
+    init(model: VoskModel, sampleRate: Float) {
+        recognizer = vosk_recognizer_new_spk(model.model, model.spkModel, sampleRate)
+    }
+    
+    deinit {
+        vosk_recognizer_free(recognizer);
+    }
+    
     func recognizeFile() -> String {
         var sres = ""
+        
         if let resourcePath = Bundle.main.resourcePath {
-
-            // Set to -1 to disable logs
-            vosk_set_log_level(0);
-
-            let modelPath = resourcePath + "/vosk-model-small-en-us-0.4"
-            let spkModelPath = resourcePath + "/vosk-model-spk-0.4"
-
-            let model = vosk_model_new(modelPath)
-            let spkModel = vosk_spk_model_new(spkModelPath)
-            let recognizer = vosk_recognizer_new_spk(model, spkModel, 16000.0)
-
+            
             let audioFile = URL(fileURLWithPath: resourcePath + "/10001-90210-01803.wav")
+    
             if let data = try? Data(contentsOf: audioFile) {
-                let _ = data.withUnsafeBytes {
-                    vosk_recognizer_accept_waveform(recognizer, $0, Int32(data.count))
-                }
-                let res = vosk_recognizer_final_result(recognizer);
-                sres = String(validatingUTF8: res!)!;
-                print(sres);
+                    let _ = data.withUnsafeBytes {
+                        vosk_recognizer_accept_waveform(recognizer, $0, Int32(data.count))
+                    }
+                    let res = vosk_recognizer_final_result(recognizer);
+                    sres = String(validatingUTF8: res!)!;
+                    print(sres);
             }
-
-            vosk_recognizer_free(recognizer)
-            vosk_model_free(model)
         }
+        
         return sres
     }
+    
+    
+    func recognizeData(buffer : AVAudioPCMBuffer) -> String {
+        let dataLen = Int(buffer.frameLength * 2)
+        let channels = UnsafeBufferPointer(start: buffer.int16ChannelData, count: 1)
+        let endOfSpeech = channels[0].withMemoryRebound(to: Int8.self, capacity: dataLen) {
+            vosk_recognizer_accept_waveform(recognizer, $0, Int32(dataLen))
+        }
+        let res = endOfSpeech == 1 ?vosk_recognizer_result(recognizer) :vosk_recognizer_partial_result(recognizer)
+        return String(validatingUTF8: res!)!;
+    }
 }

ios/VoskApiTest/VoskModel.swift

@@ -0,0 +1,36 @@
+//
+//  Vosk.swift
+//  VoskApiTest
+//
+//  Created by Niсkolay Shmyrev on 01.03.20.
+//  Copyright © 2020-2021 Alpha Cephei. All rights reserved.
+//
+
+import Foundation
+
+public final class VoskModel {
+    
+    var model : OpaquePointer!
+    var spkModel : OpaquePointer!
+    
+    init() {
+        
+        // Set to -1 to disable logs
+        vosk_set_log_level(0);
+        
+        if let resourcePath = Bundle.main.resourcePath {
+            let modelPath = resourcePath + "/vosk-model-small-en-us-0.15"
+            let spkModelPath = resourcePath + "/vosk-model-spk-0.4"
+            
+            model = vosk_model_new(modelPath)
+            spkModel = vosk_spk_model_new(spkModelPath)
+        }
+    }
+    
+    deinit {
+        vosk_model_free(model)
+        vosk_spk_model_free(spkModel)
+    }
+    
+}
+

java/Makefile

@@ -58,14 +58,14 @@ clean:
 	$(RM) -r org model-en
 
 model:
-	wget https://alphacephei.com/kaldi/models/vosk-model-small-en-us-0.3.zip
-	unzip vosk-model-small-en-us-0.3.zip && rm vosk-model-small-en-us-0.3.zip
-	mv vosk-model-small-en-us-0.3 model
+	wget https://alphacephei.com/vosk/models/vosk-model-small-en-us-0.15.zip
+	unzip vosk-model-small-en-us-0.15.zip && rm vosk-model-small-en-us-0.15.zip
+	mv vosk-model-small-en-us-0.15 model
 
 model-spk:
-	wget https://alphacephei.com/kaldi/models/vosk-model-spk-0.3.zip
-	unzip vosk-model-spk-0.3.zip && rm vosk-model-spk-0.3.zip
-	mv vosk-model-spk-0.3 model-spk
+	wget https://alphacephei.com/vosk/models/vosk-model-spk-0.4.zip
+	unzip vosk-model-spk-0.4.zip && rm vosk-model-spk-0.4.zip
+	mv vosk-model-spk-0.4 model-spk
 
 run: model model-spk
 	javac test/*.java org/kaldi/*.java

nodejs/package.json

@@ -1,6 +1,6 @@
 {
   "name": "vosk",
-  "version": "0.3.15",
+  "version": "0.3.17",
   "description": "Node binding for continuous offline voice recoginition with Vosk library.",
   "repository": {
     "type": "git",

python/example/test_microphone.py

@@ -1,28 +1,89 @@
 #!/usr/bin/env python3
 
-from vosk import Model, KaldiRecognizer
+import argparse
 import os
+import queue
+import sounddevice as sd
+import vosk
+import sys
 
-if not os.path.exists("model"):
-    print ("Please download the model from https://alphacephei.com/vosk/models and unpack as 'model' in the current folder.")
-    exit (1)
+q = queue.Queue()
 
-import pyaudio
+def int_or_str(text):
+    """Helper function for argument parsing."""
+    try:
+        return int(text)
+    except ValueError:
+        return text
 
-model = Model("model")
-rec = KaldiRecognizer(model, 16000)
+def callback(indata, frames, time, status):
+    """This is called (from a separate thread) for each audio block."""
+    if status:
+        print(status, file=sys.stderr)
+    q.put(indata)
 
-p = pyaudio.PyAudio()
-stream = p.open(format=pyaudio.paInt16, channels=1, rate=16000, input=True, frames_per_buffer=8000)
-stream.start_stream()
+parser = argparse.ArgumentParser(add_help=False)
+parser.add_argument(
+    '-l', '--list-devices', action='store_true',
+    help='show list of audio devices and exit')
+args, remaining = parser.parse_known_args()
+if args.list_devices:
+    print(sd.query_devices())
+    parser.exit(0)
+parser = argparse.ArgumentParser(
+    description=__doc__,
+    formatter_class=argparse.RawDescriptionHelpFormatter,
+    parents=[parser])
+parser.add_argument(
+    '-f', '--filename', type=str, metavar='FILENAME',
+    help='audio file to store recording to')
+parser.add_argument(
+    '-m', '--model', type=str, metavar='MODEL_PATH',
+    help='Path to the model')
+parser.add_argument(
+    '-d', '--device', type=int_or_str,
+    help='input device (numeric ID or substring)')
+parser.add_argument(
+    '-r', '--samplerate', type=int, help='sampling rate')
+args = parser.parse_args(remaining)
 
-while True:
-    data = stream.read(4000)
-    if len(data) == 0:
-        break
-    if rec.AcceptWaveform(data):
-        print(rec.Result())
+try:
+    if args.model is None:
+        args.model = "model"
+    if not os.path.exists(args.model):
+        print ("Please download a model for your language from https://alphacephei.com/vosk/models")
+        print ("and unpack as 'model' in the current folder.")
+        parser.exit(0)
+    if args.samplerate is None:
+        device_info = sd.query_devices(args.device, 'input')
+        # soundfile expects an int, sounddevice provides a float:
+        args.samplerate = int(device_info['default_samplerate'])
+
+    model = vosk.Model(args.model)
+
+    if args.filename:
+        dump_fn = open(args.filename, "wb")
     else:
-        print(rec.PartialResult())
+        dump_fn = None
 
-print(rec.FinalResult())
+    with sd.RawInputStream(samplerate=args.samplerate, blocksize = 8000, device=args.device, dtype='int16',
+                            channels=1, callback=callback):
+            print('#' * 80)
+            print('Press Ctrl+C to stop the recording')
+            print('#' * 80)
+
+            rec = vosk.KaldiRecognizer(model, args.samplerate)
+            while True:
+                data = q.get()
+                if rec.AcceptWaveform(data):
+                    print(rec.Result())
+                else:
+                    print(rec.PartialResult())
+                if dump_fn is not None:
+                    dump_fn.write(data)
+
+except KeyboardInterrupt:
+    print('\nDone')
+    parser.exit(0)
+except Exception as e:
+    parser.exit(type(e).__name__ + ': ' + str(e))

python/setup.py

@@ -1,81 +1,50 @@
 import os
 import sys
 import setuptools
-from setuptools import Extension
-from setuptools.command.build_py import build_py as _build_py
-import distutils.dir_util
-import distutils.log
+import shutil
+import glob
+import platform
 
-class build_py(_build_py):
-    def run(self):
-        self.run_command("build_ext")
-        return super().run()
+# Figure out environment for cross-compile
+vosk_source = os.getenv("VOSK_SOURCE", os.path.abspath(os.path.join(os.path.dirname(__file__), "..")))
+system = os.environ.get('VOSK_PLATFORM', platform.system())
+architecture = os.environ.get('VOSK_ARCHITECTURE', platform.architecture()[0])
 
-kaldi_root = os.getenv('KALDI_ROOT')
-kaldi_mkl = os.getenv('KALDI_MKL')
-source_path = os.getenv("VOSK_SOURCE", os.path.abspath(os.path.join(os.path.abspath(os.path.dirname(__file__)), "../src")))
+# Copy precompmilled libraries
+for lib in glob.glob(os.path.join(vosk_source, "src/lib*.*")):
+    print ("Adding library", lib)
+    shutil.copy(lib, "vosk")
 
-if kaldi_root == None:
-    print("Define KALDI_ROOT")
-    exit(1)
-
-distutils.log.set_verbosity(distutils.log.DEBUG)
-distutils.dir_util.copy_tree(
-    source_path,
-    "vosk",
-    update=1,
-    verbose=1)
+# Create OS-dependent, but Python-independent wheels.
+try:
+    from wheel.bdist_wheel import bdist_wheel
+except ImportError:
+    cmdclass = {}
+else:
+    class bdist_wheel_tag_name(bdist_wheel):
+        def get_tag(self):
+            abi = 'none'
+            if system == 'Darwin':
+                oses = 'macosx_10_6_x86_64'
+            elif system == 'Windows' and architecture == '32bit':
+                oses = 'win32'
+            elif system == 'Windows' and architecture == '64bit':
+                oses = 'win_amd64'
+            elif system == 'Linux' and architecture == '64bit':
+                oses = 'linux_x86_64'
+            elif system == 'Linux':
+                oses = 'linux_' + architecture
+            else:
+                raise TypeError("Unknown build environment")
+            return 'py3', abi, oses
+    cmdclass = {'bdist_wheel': bdist_wheel_tag_name}
 
 with open("README.md", "r") as fh:
     long_description = fh.read()
 
-kaldi_static_libs = ['src/online2/kaldi-online2.a',
-             'src/decoder/kaldi-decoder.a',
-             'src/ivector/kaldi-ivector.a',
-             'src/gmm/kaldi-gmm.a',
-             'src/nnet3/kaldi-nnet3.a',
-             'src/tree/kaldi-tree.a',
-             'src/feat/kaldi-feat.a',
-             'src/lat/kaldi-lat.a',
-             'src/lm/kaldi-lm.a',
-             'src/hmm/kaldi-hmm.a',
-             'src/transform/kaldi-transform.a',
-             'src/cudamatrix/kaldi-cudamatrix.a',
-             'src/matrix/kaldi-matrix.a',
-             'src/fstext/kaldi-fstext.a',
-             'src/util/kaldi-util.a',
-             'src/base/kaldi-base.a',
-             'tools/openfst/lib/libfst.a',
-             'tools/openfst/lib/libfstngram.a']
-kaldi_link_args = ['-s', '-latomic', '-lpthread', '-lm']
-kaldi_libraries = []
-
-if sys.platform.startswith('darwin'):
-    kaldi_link_args.extend(['-Wl,-undefined,dynamic_lookup', '-framework', 'Accelerate'])
-elif kaldi_mkl == "1":
-    kaldi_link_args.extend(['-L/opt/intel/mkl/lib/intel64', '-Wl,-rpath=/opt/intel/mkl/lib/intel64'])
-    kaldi_libraries.extend(['mkl_rt', 'mkl_intel_lp64', 'mkl_core', 'mkl_sequential'])
-else:
-    kaldi_static_libs.extend(['tools/OpenBLAS/install/lib/libopenblas.a',
-                             'tools/OpenBLAS/install/lib/liblapack.a',
-                             'tools/OpenBLAS/install/lib/libblas.a',
-                             'tools/OpenBLAS/install/lib/libf2c.a'])
-
-sources = ['kaldi_recognizer.cc', 'model.cc', 'spk_model.cc', 'vosk_api.cc', 'language_model.cc', 'vosk.i']
-
-vosk_ext = Extension('vosk._vosk',
-                    define_macros = [('FST_NO_DYNAMIC_LINKING', '1')],
-                    include_dirs = [kaldi_root + '/src', kaldi_root + '/tools/openfst/include', 'vosk'],
-                    swig_opts=['-outdir', 'vosk', '-c++'],
-                    libraries = kaldi_libraries,
-                    extra_objects = [kaldi_root + '/' + x for x in kaldi_static_libs],
-                    sources = ['vosk/' + x for x in sources],
-                    extra_link_args = kaldi_link_args,
-                    extra_compile_args = ['-O3', '-std=c++17', '-Wno-sign-compare', '-Wno-unused-variable', '-Wno-unused-local-typedefs'])
-
 setuptools.setup(
     name="vosk",
-    version="0.3.16",
+    version="0.3.18",
     author="Alpha Cephei Inc",
     author_email="contact@alphacephei.com",
     description="Offline open source speech recognition API based on Kaldi and Vosk",
@@ -83,8 +52,8 @@ setuptools.setup(
     long_description_content_type="text/markdown",
     url="https://github.com/alphacep/vosk-api",
     packages=setuptools.find_packages(),
-    ext_modules=[vosk_ext],
-    cmdclass = {'build_py' : build_py},
+    package_data = {'vosk': ['*.so', '*.dll']},
+    include_package_data=True,
     classifiers=[
         'Programming Language :: Python :: 3',
         'License :: OSI Approved :: Apache Software License',
@@ -93,5 +62,10 @@ setuptools.setup(
         'Operating System :: MacOS :: MacOS X',
         'Topic :: Software Development :: Libraries :: Python Modules'
     ],
-    python_requires='>=3.5',
+    cmdclass=cmdclass,
+    python_requires='>=3',
+    zip_safe=False, # Since we load so file from the filesystem, we can not run from zip file
+    setup_requires=['cffi>=1.0'],
+    install_requires=['cffi>=1.0'],
+    cffi_modules=['vosk_builder.py:ffibuilder'],
 )

python/vosk/__init__.py

@@ -1,2 +1,67 @@
-from .vosk import KaldiRecognizer, Model, SpkModel, SetLogLevel
+import os
+import sys
 
+from .vosk_cffi import ffi as _ffi
+
+def open_dll():
+    if sys.platform == 'win32':
+        os.environ["PATH"] += os.pathsep + os.path.dirname(__file__)
+        return _ffi.dlopen("libvosk.dll")
+    elif sys.platform == 'linux':
+        return _ffi.dlopen(os.path.dirname(__file__) + "/libvosk.so")
+    elif sys.platform == 'darwin':
+        return _ffi.dlopen(os.path.dirname(__file__) + "/libvosk.dyld")
+    else:
+        raise TypeError("Unsupported platform")
+
+_c = open_dll()
+
+class Model(object):
+
+    def __init__(self, model_path):
+        self._handle = _c.vosk_model_new(model_path.encode('utf-8'))
+
+    def __del__(self):
+        _c.vosk_model_free(self._handle)
+
+    def vosk_model_find_word(self, word):
+        return _c.vosk_model_find_word(self._handle, word.encode('utf-8'))
+
+class SpkModel(object):
+
+    def __init__(self, model_path):
+        self._handle = _c.vosk_spk_model_new(model_path.encode('utf-8'))
+
+    def __del__(self):
+        _c.vosk_spk_model_free(self._handle)
+
+class KaldiRecognizer(object):
+
+    def __init__(self, *args):
+        if len(args) == 2:
+            self._handle = _c.vosk_recognizer_new(args[0]._handle, args[1])
+        elif len(args) == 3 and type(args[1]) is SpkModel:
+            self._handle = _c.vosk_recognizer_new_spk(args[0]._handle, args[1]._handle, args[2])
+        elif len(args) == 3 and type(args[2]) is str:
+            self._handle = _c.vosk_recognizer_new_grm(args[0]._handle, args[1], args[2].encode('utf-8'))
+        else:
+            raise TypeError("Unknown arguments")
+
+    def __del__(self):
+        _c.vosk_recognizer_free(self._handle)
+
+    def AcceptWaveform(self, data):
+        return _c.vosk_recognizer_accept_waveform(self._handle, data, len(data))
+
+    def Result(self):
+        return _ffi.string(_c.vosk_recognizer_result(self._handle)).decode('utf-8')
+
+    def PartialResult(self):
+        return _ffi.string(_c.vosk_recognizer_partial_result(self._handle)).decode('utf-8')
+
+    def FinalResult(self):
+        return _ffi.string(_c.vosk_recognizer_final_result(self._handle)).decode('utf-8')
+
+
+def SetLogLevel(level):
+    return _c.vosk_set_log_level(level)

python/vosk_builder.py

@@ -0,0 +1,14 @@
+#!/usr/bin/env python3
+
+import os
+from cffi import FFI
+
+vosk_root=os.environ.get("VOSK_SOURCE", "..")
+cpp_command = "cpp " + vosk_root + "/src/vosk_api.h"
+
+ffibuilder = FFI()
+ffibuilder.set_source("vosk.vosk_cffi", None)
+ffibuilder.cdef(os.popen(cpp_command).read())
+
+if __name__ == '__main__':
+    ffibuilder.compile(verbose=True)

src/Makefile

@@ -0,0 +1,47 @@
+KALDI_ROOT?=$(HOME)/travis/kaldi
+OPENFST_ROOT?=$(KALDI_ROOT)/tools/openfst
+OPENBLAS_ROOT?=$(KALDI_ROOT)/tools/OpenBLAS/install
+EXT?=so
+CXX?=g++
+VOSK_SOURCES= \
+    kaldi_recognizer.cc \
+    language_model.cc \
+    model.cc \
+    spk_model.cc \
+    vosk_api.cc
+
+CFLAGS=-g -O2 -std=c++17 -fPIC -DFST_NO_DYNAMIC_LINKING -I. -I$(KALDI_ROOT)/src -I$(OPENFST_ROOT)/include -I$(OPENBLAS_ROOT)/include
+LIBS= \
+	$(KALDI_ROOT)/src/online2/kaldi-online2.a \
+	$(KALDI_ROOT)/src/decoder/kaldi-decoder.a \
+	$(KALDI_ROOT)/src/ivector/kaldi-ivector.a \
+	$(KALDI_ROOT)/src/gmm/kaldi-gmm.a \
+	$(KALDI_ROOT)/src/nnet3/kaldi-nnet3.a \
+	$(KALDI_ROOT)/src/tree/kaldi-tree.a \
+	$(KALDI_ROOT)/src/feat/kaldi-feat.a \
+	$(KALDI_ROOT)/src/lat/kaldi-lat.a \
+	$(KALDI_ROOT)/src/lm/kaldi-lm.a \
+	$(KALDI_ROOT)/src/hmm/kaldi-hmm.a \
+	$(KALDI_ROOT)/src/transform/kaldi-transform.a \
+	$(KALDI_ROOT)/src/cudamatrix/kaldi-cudamatrix.a \
+	$(KALDI_ROOT)/src/matrix/kaldi-matrix.a \
+	$(KALDI_ROOT)/src/fstext/kaldi-fstext.a \
+	$(KALDI_ROOT)/src/util/kaldi-util.a \
+	$(KALDI_ROOT)/src/base/kaldi-base.a \
+	$(OPENBLAS_ROOT)/lib/libopenblas.a \
+	$(OPENBLAS_ROOT)/lib/liblapack.a \
+	$(OPENBLAS_ROOT)/lib/libblas.a \
+	$(OPENBLAS_ROOT)/lib/libf2c.a \
+	$(OPENFST_ROOT)/lib/libfst.a \
+	$(OPENFST_ROOT)/lib/libfstngram.a
+
+all: libvosk.$(EXT)
+
+libvosk.$(EXT): $(VOSK_SOURCES:.cc=.o)
+	$(CXX) --shared -s -o $@ $^ $(LIBS) -lm -latomic
+
+%.o: %.cc
+	$(CXX) $(CFLAGS) -c -o $@ $<
+
+clean:
+	rm -f *.o *.so *.dll

travis/Dockerfile.dockcross

@@ -1,5 +1,5 @@
-ARG DOCKCROSS_IMAGE=linux-armv7
-FROM dockcross/${DOCKCROSS_IMAGE}
+ARG DOCKCROSS_IMAGE=alphacep/dockcross-linux-armv7
+FROM ${DOCKCROSS_IMAGE}
 
 LABEL description="A docker image for building portable Python linux binary wheels and Kaldi on other architectures"
 LABEL maintainer="contact@alphacephei.com"
@@ -14,76 +14,13 @@ RUN apt-get update && \
         autoconf \
         libtool \
         cmake \
+        python3 \
+        python3-pip \
+        python3-wheel \
+        python3-setuptools \
+        python3-cffi \
     && rm -rf /var/lib/apt/lists/*
 
-RUN cd /opt \
-    && wget -O swig-4.0.1.tar.gz https://sourceforge.net/projects/swig/files/swig/swig-4.0.1/swig-4.0.1.tar.gz/download \
-    && tar xf swig-4.0.1.tar.gz \
-    && cd swig-4.0.1 \
-    && CPP=/usr/bin/cpp CXX=/usr/bin/g++ CC=/usr/bin/gcc ./configure --prefix=/usr && make -j 10 && make install \
-    && cd .. \
-    && rm -rf swig-4.0.1.tar.gz swig-4.0.1
-
-RUN cd /opt \
-    && wget -q https://github.com/openssl/openssl/archive/OpenSSL_1_0_2u.tar.gz \
-    && tar xf OpenSSL_1_0_2u.tar.gz \
-    && cd openssl-OpenSSL_1_0_2u \
-    && CROSS_COMPILE= MACHINE="$(echo $CROSS_TRIPLE|cut -d - -f 1)" ./config --prefix=$CROSS_ROOT shared \
-    && make -j $(nproc) \
-    && make install \
-    && rm -rf /opt/openssl-OpenSSL_1_0_2u /opt/OpenSSL_1_0_2u.tar.gz
-
-RUN cd /opt \
-    && wget -q https://github.com/python/cpython/archive/v3.7.6.tar.gz \
-    && tar xf v3.7.6.tar.gz \
-    && cp -r cpython-3.7.6 cpython-3.7.6-cross \
-    && cd /opt/cpython-3.7.6 \
-    && AR=/usr/bin/ar RANLIB=/usr/bin/ranlib CPP=/usr/bin/cpp CXX=/usr/bin/g++ CC=/usr/bin/gcc  ./configure --prefix="/opt/python/cp3.7-cp3.7m" \
-    && make -j $(nproc) \
-    && make install \
-    && /opt/python/cp3.7-cp3.7m/bin/pip3 install -U pip \
-    && /opt/python/cp3.7-cp3.7m/bin/pip3 install -U wheel \
-    && cd /opt/cpython-3.7.6-cross \
-    && export PATH=/opt/python/cp3.7-cp3.7m/bin:$PATH \
-    && ./configure --prefix=$CROSS_ROOT --with-openssl=$CROSS_ROOT --host=${CROSS_TRIPLE} --build=x86-linux-gnu --disable-ipv6 ac_cv_file__dev_ptmx=no ac_cv_file__dev_ptc=no ac_cv_have_long_long_format=yes \
-    && make -j $(nproc) \
-    && make install \
-    && rm -rf /opt/cpython-3.7.6 /opt/cpython-3.7.6-cross /opt/v3.7.6.tar.gz
-
-RUN cd /opt \
-    && wget -q https://github.com/python/cpython/archive/v3.6.10.tar.gz \
-    && tar xf v3.6.10.tar.gz \
-    && cp -r cpython-3.6.10 cpython-3.6.10-cross \
-    && cd /opt/cpython-3.6.10 \
-    && AR=/usr/bin/ar RANLIB=/usr/bin/ranlib CPP=/usr/bin/cpp CXX=/usr/bin/g++ CC=/usr/bin/gcc  ./configure --prefix="/opt/python/cp3.6-cp3.6m" \
-    && make -j $(nproc) \
-    && make install \
-    && /opt/python/cp3.6-cp3.6m/bin/pip3 install -U pip \
-    && /opt/python/cp3.6-cp3.6m/bin/pip3 install -U wheel \
-    && cd /opt/cpython-3.6.10-cross \
-    && export PATH=/opt/python/cp3.6-cp3.6m/bin:$PATH \
-    && ./configure --prefix=$CROSS_ROOT --with-openssl=$CROSS_ROOT --host=${CROSS_TRIPLE} --build=x86-linux-gnu --disable-ipv6 ac_cv_file__dev_ptmx=no ac_cv_file__dev_ptc=no ac_cv_have_long_long_format=yes \
-    && make -j $(nproc) \
-    && make install \
-    && rm -rf /opt/cpython-3.6.10 /opt/cpython-3.6.10-cross /opt/v3.6.10.tar.gz
-
-RUN cd /opt \
-    && wget -q https://github.com/python/cpython/archive/v3.8.3.tar.gz \
-    && tar xf v3.8.3.tar.gz \
-    && cp -r cpython-3.8.3 cpython-3.8.3-cross \
-    && cd /opt/cpython-3.8.3 \
-    && AR=/usr/bin/ar RANLIB=/usr/bin/ranlib CPP=/usr/bin/cpp CXX=/usr/bin/g++ CC=/usr/bin/gcc  ./configure --prefix="/opt/python/cp3.8-cp3.8m" \
-    && make -j $(nproc) \
-    && make install \
-    && /opt/python/cp3.8-cp3.8m/bin/pip3 install -U pip \
-    && /opt/python/cp3.8-cp3.8m/bin/pip3 install -U wheel \
-    && cd /opt/cpython-3.8.3-cross \
-    && export PATH=/opt/python/cp3.8-cp3.8m/bin:$PATH \
-    && ./configure --prefix=$CROSS_ROOT --with-openssl=$CROSS_ROOT --host=${CROSS_TRIPLE} --build=x86-linux-gnu --disable-ipv6 ac_cv_file__dev_ptmx=no ac_cv_file__dev_ptc=no ac_cv_have_long_long_format=yes \
-    && make -j $(nproc) \
-    && make install \
-    && rm -rf /opt/cpython-3.8.3 /opt/cpython-3.8.3-cross /opt/v3.8.3.tar.gz
-
 ARG OPENBLAS_ARCH=ARMV7
 ARG ARM_HARDWARE_OPTS="-mfloat-abi=hard -mfpu=neon"
 RUN cd /opt \

travis/Dockerfile.manylinux

@@ -4,9 +4,6 @@ LABEL description="A docker image for building portable Python linux binary whee
 LABEL maintainer="contact@alphacephei.com"
 
 RUN yum -y update && yum -y install \
-    wget \
-    openssl-devel \
-    pcre-devel \
     devtoolset-8-libatomic-devel \
     automake \
     autoconf \
@@ -14,20 +11,12 @@ RUN yum -y update && yum -y install \
     cmake \
     && yum clean all
 
-RUN cd /opt \
-    && wget -O swig-4.0.1.tar.gz https://sourceforge.net/projects/swig/files/swig/swig-4.0.1/swig-4.0.1.tar.gz/download \
-    && tar xf swig-4.0.1.tar.gz \
-    && cd swig-4.0.1 \
-    && ./configure --prefix=/usr && make -j 10 && make install \
-    && cd .. \
-    && rm -rf swig-4.0.1.tar.gz swig-4.0.1
-
 RUN cd /opt \
     && git clone -b lookahead-1.8.0 --single-branch https://github.com/alphacep/kaldi \
     && cd /opt/kaldi/tools \
     && git clone -b v0.3.13 --single-branch https://github.com/xianyi/OpenBLAS \
     && git clone -b v3.2.1  --single-branch https://github.com/alphacep/clapack \
-    && make -C OpenBLAS ONLY_CBLAS=1 DYNAMIC_ARCH=1 USE_LOCKING=1 USE_THREAD=0 all \
+    && make -C OpenBLAS ONLY_CBLAS=1 DYNAMIC_ARCH=1 TARGET=NEHALEM USE_LOCKING=1 USE_THREAD=0 all \
     && make -C OpenBLAS PREFIX=$(pwd)/OpenBLAS/install install \
     && mkdir -p clapack/BUILD && cd clapack/BUILD && cmake .. && make -j 10 && find . -name "*.a" | xargs cp -t ../../OpenBLAS/install/lib \
     && cd /opt/kaldi/tools \
@@ -38,5 +27,7 @@ RUN cd /opt \
     && make -j 10 && make install \
     && cd /opt/kaldi/src \
     && ./configure --mathlib=OPENBLAS_CLAPACK --shared --use-cuda=no \
-    && make -j 10 online2 lm rnnlm \
+    && sed -i 's:-msse -msse2:-msse -msse2:g' kaldi.mk \
+    && sed -i 's: -O1 : -O3 :g' kaldi.mk \
+    && make -j $(nproc) online2 lm rnnlm \
     && find /opt/kaldi -name "*.o" -exec rm {} \;

travis/Dockerfile.win

@@ -0,0 +1,65 @@
+FROM debian:10.4
+
+RUN apt-get update && \
+    apt-get install -y --no-install-recommends \
+        ca-certificates \
+        g++ \
+        bzip2 \
+        unzip \
+        make \
+        wget \
+        git \
+        python3 \
+        python3-pip \
+        python3-wheel \
+        python3-setuptools \
+        python3-cffi \
+        zlib1g-dev \
+        patch \
+        cmake \
+        xz-utils \
+        automake \
+        autoconf \
+        libtool \
+        pkg-config \
+        sudo \
+        g++-mingw-w64-i686 \
+        g++-mingw-w64-x86-64 \
+    && rm -rf /var/lib/apt/lists/*
+
+RUN mkdir /opt/kaldi \
+    && git clone https://github.com/alphacep/openfst \
+    && cd openfst \
+    && autoreconf -i \
+    && CXX=x86_64-w64-mingw32-g++-posix CXXFLAGS="-O3 -ftree-vectorize -DFST_NO_DYNAMIC_LINKING" \
+        ./configure --prefix=/opt/kaldi/local \
+        --enable-shared --enable-static --with-pic --disable-bin \
+        --enable-lookahead-fsts --enable-ngram-fsts --host=x86_64-w64-mingw32 \
+    && make -j $(nproc) \
+    && make install
+
+RUN cd /opt/kaldi \
+    && git clone -b v0.3.13 --single-branch https://github.com/xianyi/OpenBLAS \
+    && cd OpenBLAS \
+    && make HOSTCC=gcc BINARY=64 CC=x86_64-w64-mingw32-gcc ONLY_CBLAS=1 DYNAMIC_ARCH=1 TARGET=NEHALEM USE_LOCKING=1 USE_THREAD=0 -j $(nproc) \
+    && make PREFIX=/opt/kaldi/local install
+
+RUN cd /opt/kaldi \
+    && git clone -b v3.2.1  --single-branch https://github.com/alphacep/clapack \
+    && mkdir clapack/BUILD \
+    && cd clapack/BUILD \
+    && cmake -DCMAKE_C_COMPILER_TARGET=x86_64-w64-mingw32 -DCMAKE_C_COMPILER=x86_64-w64-mingw32-gcc-posix -DCMAKE_SYSTEM_NAME=Windows -DCMAKE_CROSSCOMPILING=True .. \
+    && make -C F2CLIBS/libf2c \
+    && make -C BLAS \
+    && make -C SRC \
+    && find . -name *.a -exec cp {} /opt/kaldi/local/lib \;
+
+RUN cd /opt/kaldi \
+    && git clone -b android-mix --single-branch https://github.com/alphacep/kaldi \
+    && cd kaldi/src \
+    && CXX=x86_64-w64-mingw32-g++-posix CXXFLAGS="-O3 -ftree-vectorize -DFST_NO_DYNAMIC_LINKING" ./configure --shared --mingw=yes --use-cuda=no \
+        --mathlib=OPENBLAS_CLAPACK \
+        --host=x86_64-w64-mingw32 --openblas-clapack-root=/opt/kaldi/local \
+        --fst-root=/opt/kaldi/local --fst-version=1.8.0 \
+    && make depend -j \
+    && make -j $(nproc) online2 lm

travis/Dockerfile.win32

@@ -0,0 +1,64 @@
+FROM debian:10.4
+
+RUN apt-get update && \
+    apt-get install -y --no-install-recommends \
+        ca-certificates \
+        g++ \
+        bzip2 \
+        unzip \
+        make \
+        wget \
+        git \
+        python3 \
+        python3-pip \
+        python3-wheel \
+        python3-setuptools \
+        python3-cffi \
+        zlib1g-dev \
+        patch \
+        cmake \
+        automake \
+        autoconf \
+        libtool \
+        pkg-config \
+        sudo \
+        g++-mingw-w64-i686 \
+    && rm -rf /var/lib/apt/lists/*
+
+RUN mkdir /opt/kaldi \
+    && git clone https://github.com/alphacep/openfst \
+    && cd openfst \
+    && autoreconf -i \
+    && CXX=i686-w64-mingw32-g++-posix CXXFLAGS="-msse2 -O3 -ftree-vectorize -DFST_NO_DYNAMIC_LINKING" \
+        ./configure --prefix=/opt/kaldi/local \
+        --enable-shared --enable-static --with-pic \
+        --disable-bin --enable-lookahead-fsts --enable-ngram-fsts \
+        --host=i686-w64-mingw32 \
+    && make -j $(nproc) \
+    && make install
+
+RUN cd /opt/kaldi \
+    && git clone -b v0.3.13 --single-branch https://github.com/xianyi/OpenBLAS \
+    && cd OpenBLAS \
+    && make HOSTCC=gcc BINARY=32 CC=i686-w64-mingw32-gcc ONLY_CBLAS=1 DYNAMIC_ARCH=1 TARGET=NEHALEM USE_LOCKING=1 USE_THREAD=0 -j $(nproc) \
+    && make PREFIX=/opt/kaldi/local install
+
+RUN cd /opt/kaldi \
+    && git clone -b v3.2.1  --single-branch https://github.com/alphacep/clapack \
+    && mkdir clapack/BUILD \
+    && cd clapack/BUILD \
+    && cmake -DCMAKE_C_COMPILER_TARGET=i686-w64-mingw32 -DCMAKE_C_COMPILER=i686-w64-mingw32-gcc-posix -DCMAKE_SYSTEM_NAME=Windows -DCMAKE_CROSSCOMPILING=True .. \
+    && make -C F2CLIBS/libf2c \
+    && make -C BLAS \
+    && make -C SRC \
+    && find . -name *.a -exec cp {} /opt/kaldi/local/lib \;
+
+RUN cd /opt/kaldi \
+    && git clone -b android-mix --single-branch https://github.com/alphacep/kaldi \
+    && cd kaldi/src \
+    && CXX=i686-w64-mingw32-g++-posix CXXFLAGS="-O3 -ftree-vectorize -DFST_NO_DYNAMIC_LINKING" ./configure --shared --mingw=yes --use-cuda=no \
+        --mathlib=OPENBLAS_CLAPACK \
+        --host=i686-w64-mingw32 --openblas-clapack-root=/opt/kaldi/local \
+        --fst-root=/opt/kaldi/local --fst-version=1.8.0 \
+    && make depend -j \
+    && make -j $(nproc) online2 lm

travis/build-dockcross.sh

@@ -3,8 +3,8 @@
 set -e
 set -x
 
-docker build --build-arg="DOCKCROSS_IMAGE=linux-armv7" --build-arg="OPENBLAS_ARCH=ARMV7" --file Dockerfile.dockcross --tag alphacep/kaldi-dockcross-armv7:latest .
-docker build --build-arg="DOCKCROSS_IMAGE=linux-arm64" --build-arg="OPENBLAS_ARCH=ARMV8" --file Dockerfile.dockcross --tag alphacep/kaldi-dockcross-arm64:latest .
+docker build --build-arg="DOCKCROSS_IMAGE=alphacep/dockcross-linux-armv7" --build-arg="OPENBLAS_ARCH=ARMV7" --file Dockerfile.dockcross --tag alphacep/kaldi-dockcross-armv7:latest .
+docker build --build-arg="DOCKCROSS_IMAGE=dockcross/linux-arm64" --build-arg="OPENBLAS_ARCH=ARMV8" --file Dockerfile.dockcross --tag alphacep/kaldi-dockcross-arm64:latest .
 
 docker run --rm -v /home/shmyrev/travis/vosk-api/:/io alphacep/kaldi-dockcross-armv7 /io/travis/build-wheels-dockcross.sh
 docker run --rm -v /home/shmyrev/travis/vosk-api/:/io alphacep/kaldi-dockcross-arm64 /io/travis/build-wheels-dockcross.sh

travis/build-docker-win.sh

@@ -0,0 +1,5 @@
+#!/bin/bash
+
+set -e -x 
+docker build --file Dockerfile.win --tag alphacep/kaldi-win:latest .
+docker run --rm -v `realpath ..`:/io alphacep/kaldi-win /io/travis/build-wheels-win.sh

travis/build-docker-win32.sh

@@ -0,0 +1,5 @@
+#!/bin/bash
+
+set -e -x 
+docker build --file Dockerfile.win32 --tag alphacep/kaldi-win32:latest .
+docker run --rm -v `realpath ..`:/io alphacep/kaldi-win32 /io/travis/build-wheels-win32.sh

travis/build-docker.sh

@@ -1,7 +1,5 @@
 #!/bin/bash
 
-set -e
-set -x
-
+set -e -x 
 docker build --file Dockerfile.manylinux --tag alphacep/kaldi-manylinux:latest .
-docker run --rm -e PLAT=manylinux2010_x86_64 -v /home/shmyrev/travis/vosk-api/:/io alphacep/kaldi-manylinux /io/travis/build-wheels.sh
+docker run --rm -v `realpath ..`:/io alphacep/kaldi-manylinux /io/travis/build-wheels.sh

travis/build-wheels-dockcross.sh

@@ -1,40 +1,18 @@
 #!/bin/bash
 set -e -x
 
-ORIG_PATH=$PATH
-for pyver in 3.6 3.7 3.8; do
+cd /opt
+git clone https://github.com/alphacep/vosk-api
+cd /opt/vosk-api/src
+KALDI_ROOT=/opt/kaldi make -j $(nproc)
 
-    export KALDI_ROOT=/opt/kaldi
-    export PATH=/opt/python/cp${pyver}-cp${pyver}m/bin:$ORIG_PATH
-    export VOSK_SOURCE=/io/src
-
-    # Python 3.8 somehow changed syconfig file name
-    sysconfig_bit="m"
-    if [ $pyver == "3.8" ]; then
-         sysconfig_bit=""
-    fi
-    if [ $pyver == "3.9" ]; then
-         sysconfig_bit=""
-    fi
-
-    case $CROSS_TRIPLE in
-        *arm-*)
-            export _PYTHON_HOST_PLATFORM=linux-armv6l
-            export _PYTHON_SYSCONFIGDATA_NAME=_sysconfigdata_${sysconfig_bit}_linux_arm-linux-gnueabihf
-            ;;
-        *armv7-*)
-            export _PYTHON_HOST_PLATFORM=linux-armv7l
-            export _PYTHON_SYSCONFIGDATA_NAME=_sysconfigdata_${sysconfig_bit}_linux_arm-linux-gnueabihf
-            ;;
-        *aarch64-*)
-            export _PYTHON_HOST_PLATFORM=linux-aarch64
-            export _PYTHON_SYSCONFIGDATA_NAME=_sysconfigdata_${sysconfig_bit}_linux_aarch64-linux-gnu
-            ;;
-    esac
-    export PYTHONHOME=$CROSS_ROOT
-    export PYTHONPATH=/opt/python/cp${pyver}-cp${pyver}m/lib/python${pyver}/site-packages:/opt/python/cp${pyver}-cp${pyver}m/lib/python${pyver}/lib-dynload
-
-    rm -rf /io/python/build
-    pip${pyver} -v wheel /io/python -w /io/wheelhouse
-
-done
+export VOSK_SOURCE=/opt/vosk-api
+case $CROSS_TRIPLE in
+    *armv7-*)
+        export VOSK_ARCHITECTURE=armv7l
+        ;;
+    *aarch64-*)
+        export VOSK_ARCHITECTURE=aarch64
+        ;;
+esac
+pip3 wheel /opt/vosk-api/python --no-deps -w /io/wheelhouse

travis/build-wheels-win.sh

@@ -0,0 +1,16 @@
+#!/bin/bash
+set -e -x
+
+cd /opt
+git clone https://github.com/alphacep/vosk-api
+cd vosk-api/src
+CXX=x86_64-w64-mingw32-g++-posix EXT=dll KALDI_ROOT=/opt/kaldi/kaldi OPENFST_ROOT=/opt/kaldi/local OPENBLAS_ROOT=/opt/kaldi/local make -j $(nproc)
+
+cp /usr/lib/gcc/x86_64-w64-mingw32/*-posix/libstdc++-6.dll /opt/vosk-api/src
+cp /usr/lib/gcc/x86_64-w64-mingw32/*-posix/libgcc_s_seh-1.dll /opt/vosk-api/src
+cp /usr/x86_64-w64-mingw32/lib/libwinpthread-1.dll /opt/vosk-api/src
+
+export VOSK_SOURCE=/opt/vosk-api
+export VOSK_PLATFORM=Windows
+export VOSK_ARCHITECTURE=64bit
+python3 -m pip -v wheel /opt/vosk-api/python --no-deps -w /io/wheelhouse

travis/build-wheels-win32.sh

@@ -0,0 +1,16 @@
+#!/bin/bash
+set -e -x
+
+cd /opt
+git clone https://github.com/alphacep/vosk-api
+cd vosk-api/src
+CXX=i686-w64-mingw32-g++-posix EXT=dll KALDI_ROOT=/opt/kaldi/kaldi OPENFST_ROOT=/opt/kaldi/local OPENBLAS_ROOT=/opt/kaldi/local make -j $(nproc)
+
+cp /usr/lib/gcc/i686-w64-mingw32/*-posix/libstdc++-6.dll /opt/vosk-api/src
+cp /usr/lib/gcc/i686-w64-mingw32/*-posix/libgcc_s_sjlj-1.dll /opt/vosk-api/src
+cp /usr/i686-w64-mingw32/lib/libwinpthread-1.dll /opt/vosk-api/src
+
+export VOSK_SOURCE=/opt/vosk-api
+export VOSK_PLATFORM=Windows
+export VOSK_ARCHITECTURE=32bit
+python3 -m pip -v wheel /opt/vosk-api/python --no-deps -w /io/wheelhouse

travis/build-wheels.sh

@@ -1,17 +1,16 @@
 #!/bin/bash
 set -e -x
 
-export KALDI_ROOT=/opt/kaldi
+cd /opt
+git clone https://github.com/alphacep/vosk-api
+cd vosk-api/src
+KALDI_ROOT=/opt/kaldi OPENFST_ROOT=/opt/kaldi/tools/openfst OPENBLAS_ROOT=/opt/kaldi/tools/OpenBLAS/install make -j $(nproc)
 
-# Compile wheels
-for pypath in /opt/python/cp3[6789]*; do
-    export VOSK_SOURCE=/io/src
-    mkdir -p /opt/wheelhouse
-    rm -rf /io/python/build
-    "${pypath}/bin/pip" -v wheel /io/python -w /opt/wheelhouse
-done
+mkdir -p /opt/wheelhouse
+export VOSK_SOURCE=/opt/vosk-api
+/opt/python/cp37*/bin/pip -v wheel /opt/vosk-api/python --no-deps -w /opt/wheelhouse
 
-# Bundle external shared libraries into the wheels
 for whl in /opt/wheelhouse/*.whl; do
-    auditwheel repair "$whl" --plat $PLAT -w /io/wheelhouse/
+    cp $whl /io/wheelhouse
+    auditwheel repair "$whl" --plat manylinux2010_x86_64 -w /io/wheelhouse
 done