wehub-resource-sync
ba4be087d5
Create PR to main with cherry-pick from release / cherry-pick (push) Failing after 0s
CICD NeMo / pre-flight (push) Failing after 0s
CICD NeMo / configure (push) Has been skipped
Build, validate, and release Neural Modules / pre-flight (push) Failing after 1s
CICD NeMo / code-linting (push) Has been skipped
Build, validate, and release Neural Modules / release (push) Has been skipped
Build, validate, and release Neural Modules / release-summary (push) Has been cancelled
CICD NeMo / cicd-test-container-build (push) Has been cancelled
CICD NeMo / cicd-import-tests (push) Has been cancelled
CICD NeMo / L0_Setup_Test_Data_And_Models (push) Has been cancelled
CICD NeMo / cicd-main-unit-tests (push) Has been cancelled
CICD NeMo / cicd-main-speech (push) Has been cancelled
CICD NeMo / Nemo_CICD_Test (push) Has been cancelled
CICD NeMo / Coverage (e2e) (push) Has been cancelled
CICD NeMo / Coverage (unit-test) (push) Has been cancelled
CodeQL / Analyze (python) (push) Has been cancelled
CICD NeMo / cicd-wait-in-queue (push) Has been cancelled
51 lines
2.1 KiB
Python
51 lines
2.1 KiB
Python
# Copyright (c) 2023, NVIDIA CORPORATION & AFFILIATES. All rights reserved.
|
|
#
|
|
# Licensed under the Apache License, Version 2.0 (the "License");
|
|
# you may not use this file except in compliance with the License.
|
|
# You may obtain a copy of the License at
|
|
#
|
|
# http://www.apache.org/licenses/LICENSE-2.0
|
|
#
|
|
# Unless required by applicable law or agreed to in writing, software
|
|
# distributed under the License is distributed on an "AS IS" BASIS,
|
|
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
# See the License for the specific language governing permissions and
|
|
# limitations under the License.
|
|
|
|
"""
|
|
This file implemented unit tests for loading all pretrained AlignerModel NGC checkpoints and generating Mel-spectrograms.
|
|
The test duration breakdowns are shown below. In general, each test for a single model is ~24 seconds on an NVIDIA RTX A6000.
|
|
"""
|
|
import pytest
|
|
import torch
|
|
|
|
from nemo.collections.tts.models import AlignerModel
|
|
|
|
available_models = [model.pretrained_model_name for model in AlignerModel.list_available_models()]
|
|
|
|
|
|
@pytest.fixture(params=available_models, ids=available_models)
|
|
def pretrained_model(request, get_language_id_from_pretrained_model_name):
|
|
model_name = request.param
|
|
language_id = get_language_id_from_pretrained_model_name(model_name)
|
|
model = AlignerModel.from_pretrained(model_name=model_name)
|
|
return model, language_id
|
|
|
|
|
|
@pytest.mark.nightly
|
|
def test_inference(pretrained_model, audio_text_pair_example_english):
|
|
model, _ = pretrained_model
|
|
audio, audio_len, text_raw = audio_text_pair_example_english
|
|
|
|
# Generate mel-spectrogram
|
|
spec, spec_len = model.preprocessor(input_signal=audio, length=audio_len)
|
|
|
|
# Process text
|
|
text_normalized = model.normalizer.normalize(text_raw, punct_post_process=True)
|
|
text_tokens = model.tokenizer(text_normalized)
|
|
text = torch.tensor(text_tokens, device=spec.device).unsqueeze(0).long()
|
|
text_len = torch.tensor(len(text_tokens), device=spec.device).unsqueeze(0).long()
|
|
|
|
# Run the Aligner
|
|
_, _ = model(spec=spec, spec_len=spec_len, text=text, text_len=text_len)
|