From a780510cf5df1419a5cb4d8ffb36e4f06b4e0426 Mon Sep 17 00:00:00 2001 From: wehub-resource-sync Date: Mon, 13 Jul 2026 10:41:06 +0000 Subject: [PATCH] docs: make Chinese README the default --- README.md | 139 +++++++++++++++++++++++++++--------------------------- 1 file changed, 69 insertions(+), 70 deletions(-) diff --git a/README.md b/README.md index a7506ca..6cbb043 100644 --- a/README.md +++ b/README.md @@ -1,80 +1,79 @@ -[![Project Status: Active -- The project has reached a stable, usable state and is being actively developed.](http://www.repostatus.org/badges/latest/active.svg)](http://www.repostatus.org/#active) -[![Documentation](https://readthedocs.com/projects/nvidia-nemo/badge/?version=main)](https://docs.nvidia.com/nemo/speech/nightly/) + +> [!NOTE] +> 本文档由 WeHub 基于上游 README 翻译整理，属于社区翻译，非官方中文文档。 +> [English](./README.en.md) · [原始项目](https://github.com/NVIDIA-NeMo/Speech) · [上游 README](https://github.com/NVIDIA-NeMo/Speech/blob/HEAD/README.md) +> 原作者、版权与许可证归属以原始项目及本仓库 LICENSE 文件为准。 + +[![项目状态：活跃 —— 项目已达到稳定、可用状态，并正在积极开发中。](http://www.repostatus.org/badges/latest/active.svg)](http://www.repostatus.org/#active) +[![文档](https://readthedocs.com/projects/nvidia-nemo/badge/?version=main)](https://docs.nvidia.com/nemo/speech/nightly/) [![CodeQL](https://github.com/nvidia/nemo/actions/workflows/codeql.yml/badge.svg?branch=main&event=push)](https://github.com/nvidia/nemo/actions/workflows/codeql.yml) -[![NeMo core license and license for collections in this repo](https://img.shields.io/badge/License-Apache%202.0-brightgreen.svg)](https://github.com/NVIDIA/NeMo/blob/master/LICENSE) -[![Release version](https://badge.fury.io/py/nemo-toolkit.svg)](https://badge.fury.io/py/nemo-toolkit) -[![Python version](https://img.shields.io/pypi/pyversions/nemo-toolkit.svg)](https://badge.fury.io/py/nemo-toolkit) -[![PyPi total downloads](https://static.pepy.tech/personalized-badge/nemo-toolkit?period=total&units=international_system&left_color=grey&right_color=brightgreen&left_text=downloads)](https://pepy.tech/project/nemo-toolkit) -[![Code style: black](https://img.shields.io/badge/code%20style-black-000000.svg)](https://github.com/psf/black) +[![本仓库中 NeMo 核心许可证及 collections 许可证](https://img.shields.io/badge/License-Apache%202.0-brightgreen.svg)](https://github.com/NVIDIA/NeMo/blob/master/LICENSE) +[![发布版本](https://badge.fury.io/py/nemo-toolkit.svg)](https://badge.fury.io/py/nemo-toolkit) +[![Python 版本](https://img.shields.io/pypi/pyversions/nemo-toolkit.svg)](https://badge.fury.io/py/nemo-toolkit) +[![PyPI 总下载量](https://static.pepy.tech/personalized-badge/nemo-toolkit?period=total&units=international_system&left_color=grey&right_color=brightgreen&left_text=downloads)](https://pepy.tech/project/nemo-toolkit) +[![代码风格：black](https://img.shields.io/badge/code%20style-black-000000.svg)](https://github.com/psf/black) # **NVIDIA NeMo Speech** -Checkout our [HuggingFace🤗 collection](https://huggingface.co/collections/nvidia/nemotron-speech) for the latest open -weight checkpoints and demos! +查看我们的 [HuggingFace🤗 集合](https://huggingface.co/collections/nvidia/nemotron-speech)，获取最新的开放权重检查点（checkpoints）和演示！ -## Updates +## 更新 -> The first release of NeMo Speech after NeMo repository split is scheduled for June 2026, as the repo undergoes transformation. -> For the latest stable released version, please use [the 26.02 NGC container](https://catalog.ngc.nvidia.com/orgs/nvidia/containers/nemo?version=26.02). +> NeMo 仓库拆分后，NeMo Speech 的首个发布计划于 2026 年 6 月推出，届时本仓库将进行转型。 +> 如需最新稳定发布版本，请使用 [26.02 NGC 容器](https://catalog.ngc.nvidia.com/orgs/nvidia/containers/nemo?version=26.02). -- 2026-06: [Nemotron-3.5-ASR-Streaming-0.6B](https://huggingface.co/nvidia/nemotron-3.5-asr-streaming-0.6b) has been released with 40 languages supported, controllable latency 80ms-1s, and 240-2400 1xH100 concurrent streams. Built on cache-aware Fastconformer architecture. -- 2026-04: [Parakeet-unified-en-0.6b](https://huggingface.co/nvidia/parakeet-unified-en-0.6b) has been released with high-quality offline and streaming (with a minimum latency of 160ms) inference in one model for English language with punctuation and capitalization support. -- 2026-03: [Nemotron 3 VoiceChat](https://build.nvidia.com/nvidia/nemotron-voicechat/modelcard) is now released in Early Access. Built on the Nemotron Nano v2 LLM backbone with Nemotron speech and TTS decoder, VoiceChat delivers full-duplex, natural, interruptible conversations with low latency. Try out [the demo](https://build.nvidia.com/nvidia/nemotron-voicechat) and apply for [early access](https://developer.nvidia.com/nemotron-voicechat-early-access). -- 2026-03: [Nemotron-Speech-Streaming v2603](https://huggingface.co/nvidia/nemotron-speech-streaming-en-0.6b) has been - updated. It has been trained on a larger and more diverse corpus, resulting in lower WER across all latency modes. - Try out [the demo](https://huggingface.co/spaces/nvidia/nemotron-speech-streaming-en-0.6b) and check out - [the NIM](https://build.nvidia.com/nvidia/nemotron-asr-streaming). -- 2026-03: [MagpieTTS v2602](https://huggingface.co/nvidia/magpie_tts_multilingual_357m) has been released with support - for 9 languages(En, Es, De, Fr, Vi, It, Zh, Hi, Ja). Try out - [the demo](https://huggingface.co/nvidia/magpie_tts_multilingual_357m) and check out - [the NIM](https://build.nvidia.com/nvidia/magpie-tts-multilingual). -- 2026-01: Nemotron-Speech-Streaming was released: One checkpoint that enables users to pick their optimal point - on the latency-accuracy Pareto curve! -- 2026-01: MagpieTTS was released. -- 2026: This repo has pivoted to focus on audio, speech, and multimodal LLM. For the last NeMo release with support for more - modalities, see [v2.7.0](https://github.com/NVIDIA-NeMo/NeMo/releases/tag/v2.7.0) -- 2025-08: [Parakeet V3](https://huggingface.co/nvidia/parakeet-tdt-0.6b-v3) and - [Canary V2](https://huggingface.co/nvidia/canary-1b-v2) have been released with speech recognition and translation - support for 25 European languages. -- 2025-06: [Canary-Qwen-2.5B](https://huggingface.co/nvidia/canary-qwen-2.5b) has been released with record-setting - 5.63% WER on English Open ASR Leaderboard. +- 2026-06：[Nemotron-3.5-ASR-Streaming-0.6B](https://huggingface.co/nvidia/nemotron-3.5-asr-streaming-0.6b) 已发布，支持 40 种语言，可控延迟 80ms-1s，在 1xH100 上可支持 240-2400 路并发流。基于 cache-aware Fastconformer 架构构建。 +- 2026-04：[Parakeet-unified-en-0.6b](https://huggingface.co/nvidia/parakeet-unified-en-0.6b) 已发布，单一模型即可为英语提供高质量离线与流式（最低延迟 160ms）推理，并支持标点与大小写。 +- 2026-03：[Nemotron 3 VoiceChat](https://build.nvidia.com/nvidia/nemotron-voicechat/modelcard) 现已以 Early Access 形式发布。基于 Nemotron Nano v2 LLM 骨干网络，搭配 Nemotron speech 与 TTS 解码器，VoiceChat 可提供全双工、自然、可打断的低延迟对话。试用 [演示](https://build.nvidia.com/nvidia/nemotron-voicechat)，并申请 [early access](https://developer.nvidia.com/nemotron-voicechat-early-access). +- 2026-03：[Nemotron-Speech-Streaming v2603](https://huggingface.co/nvidia/nemotron-speech-streaming-en-0.6b) 已 + 更新。其在更大、更多样的语料上训练，各延迟模式下的 WER 均有所降低。 + 试用 [演示](https://huggingface.co/spaces/nvidia/nemotron-speech-streaming-en-0.6b)，并查看 + [NIM](https://build.nvidia.com/nvidia/nemotron-asr-streaming). +- 2026-03：[MagpieTTS v2602](https://huggingface.co/nvidia/magpie_tts_multilingual_357m) 已发布，支持 + 9 种语言（En、Es、De、Fr、Vi、It、Zh、Hi、Ja）。试用 + [演示](https://huggingface.co/nvidia/magpie_tts_multilingual_357m)，并查看 + [NIM](https://build.nvidia.com/nvidia/magpie-tts-multilingual). +- 2026-01：Nemotron-Speech-Streaming 已发布：单一检查点即可让用户在延迟-精度 Pareto 曲线上选择最优点！ +- 2026-01：MagpieTTS 已发布。 +- 2026：本仓库已转向聚焦音频、语音与多模态 LLM。如需支持更多模态的最后一个 NeMo 版本，请参阅 [v2.7.0](https://github.com/NVIDIA-NeMo/NeMo/releases/tag/v2.7.0) +- 2025-08：[Parakeet V3](https://huggingface.co/nvidia/parakeet-tdt-0.6b-v3) 与 + [Canary V2](https://huggingface.co/nvidia/canary-1b-v2) 已发布，支持 25 种欧洲语言的语音识别与翻译。 +- 2025-06：[Canary-Qwen-2.5B](https://huggingface.co/nvidia/canary-qwen-2.5b) 已发布，在 English Open ASR Leaderboard 上创下 + 5.63% WER 纪录。 -## Introduction +## 简介 -NVIDIA NeMo Speech is built for researchers and PyTorch developers working on Speech models including Automatic Speech -Recognition (ASR), Text to Speech (TTS), and Speech LLMs. It is designed to help you efficiently create, customize, and -deploy new AI models by leveraging existing code and pre-trained model checkpoints. +NVIDIA NeMo Speech 面向从事语音模型（包括自动语音识别（Automatic Speech Recognition，ASR）、文本转语音（Text to Speech，TTS）和 Speech LLM）研究的科研人员与 PyTorch 开发者。它旨在帮助您通过复用现有代码与预训练模型检查点，高效创建、定制并部署新的 AI 模型。 -For technical documentation, please see the +技术文档请参阅 [NeMo Framework User Guide](https://docs.nvidia.com/nemo/speech/nightly/). -## Requirements +## 环境要求 -NeMo Speech works with the **Python, PyTorch, and CUDA versions of your choosing**: +NeMo Speech 可与**您自选的 Python、PyTorch 和 CUDA 版本**配合使用： -- Python 3.12 or above -- PyTorch 2.7 or above (CPU, CUDA, etc. — your choice) -- NVIDIA GPU + CUDA (required for training; recommended for inference) +- Python 3.12 或更高版本 +- PyTorch 2.7 或更高版本（CPU、CUDA 等 —— 由您选择） +- NVIDIA GPU + CUDA（训练必需；推理推荐） -If you already have a Python/PyTorch/CUDA stack that satisfies those minimums, NeMo Speech installs on top of it **without replacing it**, so your existing PyTorch build is kept (see the install options below). The versions pinned in `uv.lock` and shipped in the official container — Python 3.13, PyTorch 2.12, CUDA 12.6/13.2 — are simply the combination we actively test and support. They make setup turnkey and reproducible, but they are **not** a hard requirement. +若您已有的 Python/PyTorch/CUDA 环境满足上述最低要求，NeMo Speech 可**在不替换现有环境的前提下**安装于其之上，从而保留您当前的 PyTorch 构建（安装选项见下文）。`uv.lock` 中固定的版本，以及官方容器中提供的版本 —— Python 3.13、PyTorch 2.12、CUDA 12.6/13.2 —— 只是我们积极测试与支持的组合。它们便于开箱即用与可复现部署，但**并非**硬性要求。 -As of [Pytorch 2.6](https://docs.pytorch.org/docs/stable/notes/serialization.html#torch-load-with-weights-only-true), -`torch.load` defaults to using `weights_only=True`. Some model checkpoints may require using `weights_only=False`. -In this case, you can set the env var `TORCH_FORCE_NO_WEIGHTS_ONLY_LOAD=1` before running code that uses `torch.load`. -However, this should only be done with trusted files. Loading files from untrusted sources with more than weights only -can have the risk of arbitrary code execution. +自 [Pytorch 2.6](https://docs.pytorch.org/docs/stable/notes/serialization.html#torch-load-with-weights-only-true), +`torch.load` 默认使用 `weights_only=True`。部分模型检查点可能需要使用 `weights_only=False`。 +此时，可在运行使用 `torch.load` 的代码前设置环境变量 `TORCH_FORCE_NO_WEIGHTS_ONLY_LOAD=1`。 +但这应仅用于可信文件。从不信任来源加载不仅含权重的内容，存在任意代码执行风险。 -## Developer Documentation +## 开发者文档 -| Version | Status | Description | +| 版本 | 状态 | 描述 | | ------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------ | -| Latest | [![Documentation Status](https://readthedocs.com/projects/nvidia-nemo/badge/?version=main)](https://docs.nvidia.com/deeplearning/nemo/user-guide/docs/en/main/) | [Documentation of the latest (i.e. main) branch.](https://docs.nvidia.com/nemo/speech/nightly/) | -| Stable | [![Documentation Status](https://readthedocs.com/projects/nvidia-nemo/badge/?version=stable)](https://docs.nvidia.com/deeplearning/nemo/user-guide/docs/en/stable/) | Documentation of the stable (i.e. most recent release) - To be added | +| 最新 | [![文档状态](https://readthedocs.com/projects/nvidia-nemo/badge/?version=main)](https://docs.nvidia.com/deeplearning/nemo/user-guide/docs/en/main/) | [最新（即 main）分支的文档。](https://docs.nvidia.com/nemo/speech/nightly/) | +| 稳定 | [![文档状态](https://readthedocs.com/projects/nvidia-nemo/badge/?version=stable)](https://docs.nvidia.com/deeplearning/nemo/user-guide/docs/en/stable/) | 稳定版（即最近发布版本）的文档 —— 待补充 | -## Install NeMo Speech +## 安装 NeMo Speech -The recommended way to install NeMo Speech is from source with [uv](https://docs.astral.sh/uv/), which reproduces our actively-tested stack from the committed `uv.lock`. If you need different Python/PyTorch/CUDA versions, NeMo also installs over your existing environment via pip — see the [pip fallback](#from-pypi-with-pip-fallback--bring-your-own-versions) below. +推荐通过 [uv](https://docs.astral.sh/uv/), 从源码安装 NeMo Speech，以复现我们基于已提交的 `uv.lock` 积极测试的环境栈。若您需要不同的 Python/PyTorch/CUDA 版本，也可通过 pip 在现有环境上安装 NeMo —— 见下方 [pip 备选方案](#from-pypi-with-pip-fallback--bring-your-own-versions)。 -### From source with uv (recommended) +### 使用 uv 从源码安装（推荐） ```bash git clone https://github.com/NVIDIA-NeMo/NeMo.git @@ -82,15 +81,15 @@ cd NeMo uv sync --extra all --extra cu13 # CUDA 13.x (recommended) — use --extra cu12 for CUDA 12.x ``` -This installs our supported stack (Python 3.13, PyTorch 2.12, CUDA 13.2) into `.venv/` with NeMo editable. Add `--group test` for the test suite or `--group docs` to build the docs; run tools via `uv run ` or activate with `source .venv/bin/activate`. On Linux, `cu12` and `cu13` are mutually exclusive — pass exactly one (`cu13` is the default). For the **exact** container baseline, add `--locked --python 3.13` (the path the Dockerfile and CI use). +这会将我们支持的环境栈（Python 3.13、PyTorch 2.12、CUDA 13.2）安装到 `.venv/`，并以可编辑模式安装 NeMo。添加 `--group test` 以安装测试套件，或添加 `--group docs` 以构建文档；通过 `uv run ` 运行工具，或使用 `source .venv/bin/activate` 激活环境。在 Linux 上，`cu12` 与 `cu13` 互斥 —— 请只传入其中一个（默认是 `cu13`）。若要使用**与容器完全一致**的基线，请添加 `--locked --python 3.13`（Dockerfile 与 CI 使用的路径）。 -> **SpeechLM2 / Automodel:** the Automodel backend runs **without** any compiled dependencies. It can *optionally* benefit from dedicated accelerated backends (Transformer Engine, FlashAttention, Mamba, grouped-GEMM/MoE, DeepEP) for better performance — these source-built kernels come from the `compiled` (Hopper/Blackwell) or `compiled-a100` (A100) extras, built by `docker/Dockerfile` (`GPU_TARGET=h100plus` / `a100`). See the [installation guide](https://docs.nvidia.com/nemo/speech/nightly/) for the full list and build details. +> **SpeechLM2 / Automodel：**Automodel 后端**无需**任何编译依赖即可运行。它*可选地*受益于专用加速后端（Transformer Engine、FlashAttention、Mamba、grouped-GEMM/MoE、DeepEP）以获得更好性能——这些源码构建的内核来自 `compiled`（Hopper/Blackwell）或 `compiled-a100`（A100）附加组件，由 `docker/Dockerfile`（`GPU_TARGET=h100plus` / `a100`）构建。完整列表与构建细节请参阅[安装指南](https://docs.nvidia.com/nemo/speech/nightly/) for the full list and build details. -### Docker (turnkey, our supported stack) +### Docker（开箱即用，我们支持的栈） -> **NGC container:** _Coming soon — the pull command for the prebuilt NeMo Speech container image will be published here._ +> **NGC 容器：**_即将推出——预构建 NeMo Speech 容器镜像的拉取命令将在此发布。_ -To build the container from source (CUDA 13 / H100+ by default): +从源码构建容器（默认 CUDA 13 / H100+）： ```bash git clone https://github.com/NVIDIA-NeMo/NeMo.git @@ -99,30 +98,30 @@ docker buildx build -f docker/Dockerfile -t nemo-speech . # CUDA 13 / H docker run --rm -it --gpus all -v "$PWD:/workspace" nemo-speech bash ``` -For A100, set `GPU_TARGET=a100` — A100 works with **both CUDA 12 and CUDA 13** (CUDA 13, the default base image, is recommended; the CUDA 12 base is a convenience). See the header of [`docker/Dockerfile`](docker/Dockerfile) for all build arguments (`BASE_IMAGE`, `GPU_TARGET`). +对于 A100，请设置 `GPU_TARGET=a100`——A100 同时支持 **CUDA 12 与 CUDA 13**（推荐使用默认基础镜像 CUDA 13；CUDA 12 基础镜像仅为便利选项）。所有构建参数（`BASE_IMAGE`、`GPU_TARGET`）请参阅 [`docker/Dockerfile`](docker/Dockerfile) 文件头部。 -### From PyPI with pip (fallback — bring your own versions) +### 通过 pip 从 PyPI 安装（备选方案——自带版本） -Prefer your own Python/PyTorch/CUDA? Install your PyTorch first (any version ≥ 2.7 for your CPU/CUDA/etc. target — see the [PyTorch install matrix](https://pytorch.org/get-started/locally/)), then add NeMo and it **keeps your build**. `uv pip` (uv's fast, pip-compatible installer) works like `pip`: +想使用自己的 Python/PyTorch/CUDA？请先安装 PyTorch（针对你的 CPU/CUDA 等目标，任意 ≥ 2.7 的版本——参见 [PyTorch 安装矩阵](https://pytorch.org/get-started/locally/)), then add NeMo and it **keeps your build**. `uv pip`（uv 的快速、与 pip 兼容的安装器）的使用方式与 `pip` 相同： ```bash uv pip install 'nemo-toolkit[asr,tts]' # or plain: pip install 'nemo-toolkit[asr,tts]' ``` -> ⚠️ Do **not** use `uv sync --locked` for a bring-your-own stack — it applies `uv.lock` and replaces your Python/PyTorch/CUDA with the supported baseline. Use `uv pip`/`pip` here; reserve `uv sync --locked` for reproducing our stack. +> ⚠️ 自带技术栈时**请勿**使用 `uv sync --locked`——它会应用 `uv.lock`，并将你的 Python/PyTorch/CUDA 替换为受支持的基线版本。此处请使用 `uv pip`/`pip`；`uv sync --locked` 留作复现我们的技术栈。 -To instead pull *our* pinned PyTorch build, add the CUDA extra and the matching wheel index (pip/uv pip do not read uv's project index config, so `--extra-index-url` is required): +若要改为拉取*我们*固定的 PyTorch 构建，请添加 CUDA 额外依赖及匹配的 wheel 索引（pip/uv pip 不会读取 uv 的项目索引配置，因此需要 `--extra-index-url`）： ```bash pip install 'nemo-toolkit[asr,tts,cu13]' --extra-index-url https://download.pytorch.org/whl/cu132 # CUDA 13.x pip install 'nemo-toolkit[asr,tts,cu12]' --extra-index-url https://download.pytorch.org/whl/cu126 # CUDA 12.x ``` -## Contribute to NeMo +## 为 NeMo 做贡献 -We welcome community contributions! Please refer to +我们欢迎社区贡献！流程请参阅 [CONTRIBUTING.md](https://github.com/NVIDIA-NeMo/NeMo/blob/main/CONTRIBUTING.md) for the process. -## Licenses +## 许可证 -NeMo is licensed under the [Apache License 2.0](https://github.com/NVIDIA/NeMo?tab=Apache-2.0-1-ov-file). +NeMo 采用 [Apache License 2.0](https://github.com/NVIDIA/NeMo?tab=Apache-2.0-1-ov-file). 许可。