From 1e5395b0f31d6f38b17e6cebe7110db63fc8bdd1 Mon Sep 17 00:00:00 2001 From: 游雁 <zhifu.gzf@alibaba-inc.com> Date: 星期四, 19 十月 2023 14:00:54 +0800 Subject: [PATCH] docs --- egs_modelscope/asr/TEMPLATE/README.md | 22 ++++++++++++++++++++++ egs_modelscope/asr/TEMPLATE/README_zh.md | 22 ++++++++++++++++++++++ 2 files changed, 44 insertions(+), 0 deletions(-) diff --git a/egs_modelscope/asr/TEMPLATE/README.md b/egs_modelscope/asr/TEMPLATE/README.md index ac73950..10ddc1e 100644 --- a/egs_modelscope/asr/TEMPLATE/README.md +++ b/egs_modelscope/asr/TEMPLATE/README.md @@ -21,6 +21,28 @@ rec_result = inference_pipeline(audio_in='https://isv-data.oss-cn-hangzhou.aliyuncs.com/ics/MaaS/ASR/test_audio/asr_example_zh.wav') print(rec_result) ``` +#### [Paraformer-long Model](https://www.modelscope.cn/models/damo/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-pytorch/summary) +```python +from modelscope.pipelines import pipeline +from modelscope.utils.constant import Tasks + +inference_pipeline = pipeline( + task=Tasks.auto_speech_recognition, + model='damo/speech_paraformer-large-vad-punc_asr_nat-zh-cn-16k-common-vocab8404-pytorch', + vad_model='damo/speech_fsmn_vad_zh-cn-16k-common-pytorch', + #punc_model='damo/punc_ct-transformer_zh-cn-common-vocab272727-pytorch', + punc_model='damo/punc_ct-transformer_cn-en-common-vocab471067-large', +) + +rec_result = inference_pipeline(audio_in='https://isv-data.oss-cn-hangzhou.aliyuncs.com/ics/MaaS/ASR/test_audio/vad_example.wav', + batch_size_token=5000, batch_size_token_threshold_s=40, max_single_segment_time=6000) +print(rec_result) +``` +Where, +- `batch_size_token` refs to dynamic batch_size and the total tokens of batch is `batch_size_token`, 1 token = 60 ms. +- `batch_size_token_threshold_s`: The batch_size is set to 1, when the audio duration exceeds the threshold value of `batch_size_token_threshold_s`, specified in `s`. +- `max_single_segment_time`: The maximum length for audio segmentation in VAD, specified in `ms`. + #### [Paraformer-online Model](https://www.modelscope.cn/models/damo/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-online/summary) ##### Streaming Decoding ```python diff --git a/egs_modelscope/asr/TEMPLATE/README_zh.md b/egs_modelscope/asr/TEMPLATE/README_zh.md index 47656b3..591ac17 100644 --- a/egs_modelscope/asr/TEMPLATE/README_zh.md +++ b/egs_modelscope/asr/TEMPLATE/README_zh.md @@ -21,6 +21,28 @@ rec_result = inference_pipeline(audio_in='https://isv-data.oss-cn-hangzhou.aliyuncs.com/ics/MaaS/ASR/test_audio/asr_example_zh.wav') print(rec_result) ``` +#### [Paraformer-long Model](https://www.modelscope.cn/models/damo/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-pytorch/summary) +```python +from modelscope.pipelines import pipeline +from modelscope.utils.constant import Tasks + +inference_pipeline = pipeline( + task=Tasks.auto_speech_recognition, + model='damo/speech_paraformer-large-vad-punc_asr_nat-zh-cn-16k-common-vocab8404-pytorch', + vad_model='damo/speech_fsmn_vad_zh-cn-16k-common-pytorch', + #punc_model='damo/punc_ct-transformer_zh-cn-common-vocab272727-pytorch', + punc_model='damo/punc_ct-transformer_cn-en-common-vocab471067-large', +) + +rec_result = inference_pipeline(audio_in='https://isv-data.oss-cn-hangzhou.aliyuncs.com/ics/MaaS/ASR/test_audio/vad_example.wav', + batch_size_token=5000, batch_size_token_threshold_s=40, max_single_segment_time=6000) +print(rec_result) +``` +Where, +- `batch_size_token` 琛ㄧず閲囩敤鍔ㄦ�乥atch锛宐atch涓�籺oken鏁颁负 `batch_size_token`锛�1 token = 60 ms. +- `batch_size_token_threshold_s`: 琛ㄧず闊抽鏃堕暱瓒呰繃 `batch_size_token_threshold_s`闃堝�兼槸锛宐atch鏁拌缃负1, 鍗曚綅涓簊. +- `max_single_segment_time`: 琛ㄧずVAD鏈�澶у垏鍓查煶棰戞椂闀�, 鍗曚綅鏄痬s. + #### [Paraformer-瀹炴椂妯″瀷](https://www.modelscope.cn/models/damo/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-online/summary) ##### 瀹炴椂鎺ㄧ悊 ```python -- Gitblit v1.9.1