From 81fb78286f6e6893ef5a319bfb2ba21d340476d3 Mon Sep 17 00:00:00 2001
From: 游雁 <zhifu.gzf@alibaba-inc.com>
Date: 星期五, 22 三月 2024 20:13:05 +0800
Subject: [PATCH] update
---
README.md | 177 +++++++++++++++++++++++++++++++++++++++--------------------
1 files changed, 117 insertions(+), 60 deletions(-)
diff --git a/README.md b/README.md
index 50ca183..2409fe5 100644
--- a/README.md
+++ b/README.md
@@ -3,11 +3,10 @@
([绠�浣撲腑鏂嘳(./README_zh.md)|English)
# FunASR: A Fundamental End-to-End Speech Recognition Toolkit
-<p align="left">
- <a href=""><img src="https://img.shields.io/badge/OS-Linux%2C%20Win%2C%20Mac-brightgreen.svg"></a>
- <a href=""><img src="https://img.shields.io/badge/Python->=3.7,<=3.10-aff.svg"></a>
- <a href=""><img src="https://img.shields.io/badge/Pytorch-%3E%3D1.11-blue"></a>
-</p>
+
+
+[](https://pypi.org/project/funasr/)
+
<strong>FunASR</strong> hopes to build a bridge between academic research and industrial applications on speech recognition. By supporting the training & finetuning of the industrial-grade speech recognition model, researchers and developers can conduct research and production of speech recognition models more conveniently, and promote the development of speech recognition ecology. ASR for Fun锛�
@@ -15,6 +14,7 @@
| [**News**](https://github.com/alibaba-damo-academy/FunASR#whats-new)
| [**Installation**](#installation)
| [**Quick Start**](#quick-start)
+| [**Tutorial**](https://github.com/alibaba-damo-academy/FunASR/blob/main/docs/tutorial/README.md)
| [**Runtime**](./runtime/readme.md)
| [**Model Zoo**](#model-zoo)
| [**Contact**](#contact)
@@ -28,6 +28,12 @@
<a name="whats-new"></a>
## What's new:
+- 2024/03/05锛欰dded the Qwen-Audio and Qwen-Audio-Chat large-scale audio-text multimodal models, which have topped multiple audio domain leaderboards. These models support speech dialogue, [usage](examples/industrial_data_pretraining/qwen_audio).
+- 2024/03/05锛欰dded support for the Whisper-large-v3 model, a multitasking model that can perform multilingual speech recognition, speech translation, and language identification. It can be downloaded from the[modelscope](examples/industrial_data_pretraining/whisper/demo.py), and [openai](examples/industrial_data_pretraining/whisper/demo_from_openai.py).
+- 2024/03/05: Offline File Transcription Service 4.4, Offline File Transcription Service of English 1.5锛孯eal-time Transcription Service 1.9 released锛宒ocker image supports ARM64 platform, update modelscope锛�([docs](runtime/readme.md))
+- 2024/01/30锛歠unasr-1.0 has been released ([docs](https://github.com/alibaba-damo-academy/FunASR/discussions/1319))
+- 2024/01/30锛歟motion recognition models are new supported. [model link](https://www.modelscope.cn/models/iic/emotion2vec_base_finetuned/summary), modified from [repo](https://github.com/ddlBoJack/emotion2vec).
+- 2024/01/25: Offline File Transcription Service 4.2, Offline File Transcription Service of English 1.3 released锛宱ptimized the VAD (Voice Activity Detection) data processing method, significantly reducing peak memory usage, memory leak optimization; Real-time Transcription Service 1.7 released锛宱ptimizatized the client-side锛�([docs](runtime/readme.md))
- 2024/01/09: The Funasr SDK for Windows version 2.0 has been released, featuring support for The offline file transcription service (CPU) of Mandarin 4.1, The offline file transcription service (CPU) of English 1.2, The real-time transcription service (CPU) of Mandarin 1.6. For more details, please refer to the official documentation or release notes([FunASR-Runtime-Windows](https://www.modelscope.cn/models/damo/funasr-runtime-win-cpu-x64/summary))
- 2024/01/03: File Transcription Service 4.0 released, Added support for 8k models, optimized timestamp mismatch issues and added sentence-level timestamps, improved the effectiveness of English word FST hotwords, supported automated configuration of thread parameters, and fixed known crash issues as well as memory leak problems, refer to ([docs](runtime/readme.md#file-transcription-service-mandarin-cpu)).
- 2024/01/03: Real-time Transcription Service 1.6 released锛孴he 2pass-offline mode supports Ngram language model decoding and WFST hotwords, while also addressing known crash issues and memory leak problems, ([docs](runtime/readme.md#the-real-time-transcription-service-mandarin-cpu))
@@ -47,25 +53,40 @@
<a name="Installation"></a>
## Installation
-Please ref to [installation docs](https://alibaba-damo-academy.github.io/FunASR/en/installation/installation.html)
+```shell
+pip3 install -U funasr
+```
+Or install from source code
+``` sh
+git clone https://github.com/alibaba/FunASR.git && cd FunASR
+pip3 install -e ./
+```
+Install modelscope for the pretrained models (Optional)
+
+```shell
+pip3 install -U modelscope
+```
## Model Zoo
-FunASR has open-sourced a large number of pre-trained models on industrial data. You are free to use, copy, modify, and share FunASR models under the [Model License Agreement](./MODEL_LICENSE). Below are some representative models, for more models please refer to the [Model Zoo]().
+FunASR has open-sourced a large number of pre-trained models on industrial data. You are free to use, copy, modify, and share FunASR models under the [Model License Agreement](./MODEL_LICENSE). Below are some representative models, for more models please refer to the [Model Zoo](./model_zoo).
-(Note: 馃 represents the Huggingface model zoo link, 猸� represents the ModelScope model zoo link)
+(Note: 猸� represents the ModelScope model zoo, 馃 represents the Huggingface model zoo, 馃崁 represents the OpenAI model zoo)
-| Model Name | Task Details | Training Data | Parameters |
-|:------------------------------------------------------------------------------------------------------------------------------------------------------------------:|:---------------------------------------------------------------------------:|:--------------------------------:|:----------:|
-| paraformer-zh <br> ([猸怾(https://www.modelscope.cn/models/damo/speech_paraformer-large-vad-punc_asr_nat-zh-cn-16k-common-vocab8404-pytorch/summary) [馃]() ) | speech recognition, with timestamps, non-streaming | 60000 hours, Mandarin | 220M |
-| paraformer-zh-spk <br> ( [猸怾(https://modelscope.cn/models/damo/speech_paraformer-large-vad-punc-spk_asr_nat-zh-cn/summary) [馃]() ) | speech recognition with speaker diarization, with timestamps, non-streaming | 60000 hours, Mandarin | 220M |
-| <nobr>paraformer-zh-online <br> ( [猸怾(https://modelscope.cn/models/damo/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-online/summary) [馃]() )</nobr> | speech recognition, streaming | 60000 hours, Mandarin | 220M |
-| paraformer-en <br> ( [猸怾(https://www.modelscope.cn/models/damo/speech_paraformer-large-vad-punc_asr_nat-en-16k-common-vocab10020/summary) [馃]() ) | speech recognition, with timestamps, non-streaming | 50000 hours, English | 220M |
-| paraformer-en-spk <br> ([猸怾()[馃]() ) | speech recognition with speaker diarization, non-streaming | Undo | Undo |
-| conformer-en <br> ( [猸怾(https://modelscope.cn/models/damo/speech_conformer_asr-en-16k-vocab4199-pytorch/summary) [馃]() ) | speech recognition, non-streaming | 50000 hours, English | 220M |
-| ct-punc <br> ( [猸怾(https://modelscope.cn/models/damo/punc_ct-transformer_cn-en-common-vocab471067-large/summary) [馃]() ) | punctuation restoration | 100M, Mandarin and English | 1.1G |
-| fsmn-vad <br> ( [猸怾(https://modelscope.cn/models/damo/speech_fsmn_vad_zh-cn-16k-common-pytorch/summary) [馃]() ) | voice activity detection | 5000 hours, Mandarin and English | 0.4M |
-| fa-zh <br> ( [猸怾(https://modelscope.cn/models/damo/speech_timestamp_prediction-v1-16k-offline/summary) [馃]() ) | timestamp prediction | 5000 hours, Mandarin | 38M |
+| Model Name | Task Details | Training Data | Parameters |
+|:--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------:|:-----------------------------------------------------:|:--------------------------------:|:----------:|
+| paraformer-zh <br> ([猸怾(https://www.modelscope.cn/models/damo/speech_paraformer-large-vad-punc_asr_nat-zh-cn-16k-common-vocab8404-pytorch/summary) [馃](https://huggingface.co/funasr/paraformer-tp) ) | speech recognition, with timestamps, non-streaming | 60000 hours, Mandarin | 220M |
+| <nobr>paraformer-zh-streaming <br> ( [猸怾(https://modelscope.cn/models/damo/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-online/summary) [馃](https://huggingface.co/funasr/paraformer-zh-streaming) )</nobr> | speech recognition, streaming | 60000 hours, Mandarin | 220M |
+| paraformer-en <br> ( [猸怾(https://www.modelscope.cn/models/damo/speech_paraformer-large-vad-punc_asr_nat-en-16k-common-vocab10020/summary) [馃](https://huggingface.co/funasr/paraformer-en) ) | speech recognition, without timestamps, non-streaming | 50000 hours, English | 220M |
+| conformer-en <br> ( [猸怾(https://modelscope.cn/models/damo/speech_conformer_asr-en-16k-vocab4199-pytorch/summary) [馃](https://huggingface.co/funasr/conformer-en) ) | speech recognition, non-streaming | 50000 hours, English | 220M |
+| ct-punc <br> ( [猸怾(https://modelscope.cn/models/damo/punc_ct-transformer_cn-en-common-vocab471067-large/summary) [馃](https://huggingface.co/funasr/ct-punc) ) | punctuation restoration | 100M, Mandarin and English | 1.1G |
+| fsmn-vad <br> ( [猸怾(https://modelscope.cn/models/damo/speech_fsmn_vad_zh-cn-16k-common-pytorch/summary) [馃](https://huggingface.co/funasr/fsmn-vad) ) | voice activity detection | 5000 hours, Mandarin and English | 0.4M |
+| fa-zh <br> ( [猸怾(https://modelscope.cn/models/damo/speech_timestamp_prediction-v1-16k-offline/summary) [馃](https://huggingface.co/funasr/fa-zh) ) | timestamp prediction | 5000 hours, Mandarin | 38M |
+| cam++ <br> ( [猸怾(https://modelscope.cn/models/iic/speech_campplus_sv_zh-cn_16k-common/summary) [馃](https://huggingface.co/funasr/campplus) ) | speaker verification/diarization | 5000 hours | 7.2M |
+| Whisper-large-v2 <br> ([猸怾(https://www.modelscope.cn/models/iic/speech_whisper-large_asr_multilingual/summary) [馃崁](https://github.com/openai/whisper) ) | speech recognition, with timestamps, non-streaming | multilingual | 1.5G |
+| Whisper-large-v3 <br> ([猸怾(https://www.modelscope.cn/models/iic/Whisper-large-v3/summary) [馃崁](https://github.com/openai/whisper) ) | speech recognition, with timestamps, non-streaming | multilingual | 1.5G |
+| Qwen-Audio <br> ([猸怾(examples/industrial_data_pretraining/qwen_audio/demo.py) [馃](https://huggingface.co/Qwen/Qwen-Audio) ) | audio-text multimodal models (pretraining) | multilingual | 8B |
+| Qwen-Audio-Chat <br> ([猸怾(examples/industrial_data_pretraining/qwen_audio/demo_chat.py) [馃](https://huggingface.co/Qwen/Qwen-Audio-Chat) ) | audio-text multimodal models (chat) | multilingual | 8B |
@@ -77,12 +98,12 @@
<a name="quick-start"></a>
## Quick Start
-Below is a quick start tutorial. Test audio files ([Mandarin](https://isv-data.oss-cn-hangzhou.aliyuncs.com/ics/MaaS/ASR/test_audio/vad_example.wav), [English]()).
+Below is a quick start tutorial. Test audio files ([Mandarin](https://isv-data.oss-cn-hangzhou.aliyuncs.com/ics/MaaS/ASR/test_audio/vad_example.wav), [English](https://isv-data.oss-cn-hangzhou.aliyuncs.com/ics/MaaS/ASR/test_audio/asr_example_en.wav)).
### Command-line usage
```shell
-funasr +model=paraformer-zh +vad_model="fsmn-vad" +punc_model="ct-punc" +input=asr_example_zh.wav
+funasr ++model=paraformer-zh ++vad_model="fsmn-vad" ++punc_model="ct-punc" ++input=asr_example_zh.wav
```
Notes: Support recognition of single audio file, as well as file list in Kaldi-style wav.scp format: `wav_id wav_pat`
@@ -90,15 +111,17 @@
### Speech Recognition (Non-streaming)
```python
from funasr import AutoModel
-
-model = AutoModel(model="paraformer-zh")
-# for the long duration wav, you could add vad model
-# model = AutoModel(model="paraformer-zh", vad_model="fsmn-vad", punc_model="ct-punc")
-
-res = model(input="asr_example_zh.wav", batch_size=64)
+# paraformer-zh is a multi-functional asr model
+# use vad, punc, spk or not as you need
+model = AutoModel(model="paraformer-zh", vad_model="fsmn-vad", punc_model="ct-punc",
+ # spk_model="cam++",
+ )
+res = model.generate(input=f"{model.model_path}/example/asr_example.wav",
+ batch_size_s=300,
+ hotword='榄旀惌')
print(res)
```
-Note: `model_hub`: represents the model repository, `ms` stands for selecting ModelScope download, `hf` stands for selecting Huggingface download.
+Note: `hub`: represents the model repository, `ms` stands for selecting ModelScope download, `hf` stands for selecting Huggingface download.
### Speech Recognition (Streaming)
```python
@@ -108,7 +131,7 @@
encoder_chunk_look_back = 4 #number of chunks to lookback for encoder self-attention
decoder_chunk_look_back = 1 #number of encoder chunks to lookback for decoder cross-attention
-model = AutoModel(model="paraformer-zh-streaming", model_revision="v2.0.0")
+model = AutoModel(model="paraformer-zh-streaming")
import soundfile
import os
@@ -122,33 +145,28 @@
for i in range(total_chunk_num):
speech_chunk = speech[i*chunk_stride:(i+1)*chunk_stride]
is_final = i == total_chunk_num - 1
- res = model(input=speech_chunk,
- cache=cache,
- is_final=is_final,
- chunk_size=chunk_size,
- encoder_chunk_look_back=encoder_chunk_look_back,
- decoder_chunk_look_back=decoder_chunk_look_back,
- )
+ res = model.generate(input=speech_chunk, cache=cache, is_final=is_final, chunk_size=chunk_size, encoder_chunk_look_back=encoder_chunk_look_back, decoder_chunk_look_back=decoder_chunk_look_back)
print(res)
```
Note: `chunk_size` is the configuration for streaming latency.` [0,10,5]` indicates that the real-time display granularity is `10*60=600ms`, and the lookahead information is `5*60=300ms`. Each inference input is `600ms` (sample points are `16000*0.6=960`), and the output is the corresponding text. For the last speech segment input, `is_final=True` needs to be set to force the output of the last word.
-### Voice Activity Detection (streaming)
+### Voice Activity Detection (Non-Streaming)
```python
from funasr import AutoModel
-model = AutoModel(model="fsmn-vad", model_revision="v2.0.2")
-
+model = AutoModel(model="fsmn-vad")
wav_file = f"{model.model_path}/example/asr_example.wav"
-res = model(input=wav_file)
+res = model.generate(input=wav_file)
print(res)
```
-### Voice Activity Detection (Non-streaming)
+Note: The output format of the VAD model is: `[[beg1, end1], [beg2, end2], ..., [begN, endN]]`, where `begN/endN` indicates the starting/ending point of the `N-th` valid audio segment, measured in milliseconds.
+
+### Voice Activity Detection (Streaming)
```python
from funasr import AutoModel
chunk_size = 200 # ms
-model = AutoModel(model="fsmn-vad", model_revision="v2.0.2")
+model = AutoModel(model="fsmn-vad")
import soundfile
@@ -161,36 +179,69 @@
for i in range(total_chunk_num):
speech_chunk = speech[i*chunk_stride:(i+1)*chunk_stride]
is_final = i == total_chunk_num - 1
- res = model(input=speech_chunk,
- cache=cache,
- is_final=is_final,
- chunk_size=chunk_size,
- )
+ res = model.generate(input=speech_chunk, cache=cache, is_final=is_final, chunk_size=chunk_size)
if len(res[0]["value"]):
print(res)
```
+Note: The output format for the streaming VAD model can be one of four scenarios:
+- `[[beg1, end1], [beg2, end2], .., [begN, endN]]`锛歍he same as the offline VAD output result mentioned above.
+- `[[beg, -1]]`锛欼ndicates that only a starting point has been detected.
+- `[[-1, end]]`锛欼ndicates that only an ending point has been detected.
+- `[]`锛欼ndicates that neither a starting point nor an ending point has been detected.
+
+The output is measured in milliseconds and represents the absolute time from the starting point.
### Punctuation Restoration
```python
from funasr import AutoModel
-model = AutoModel(model="ct-punc", model_revision="v2.0.1")
-
-res = model(input="閭ｄ粖澶╃殑浼氬氨鍒拌繖閲屽惂 happy new year 鏄庡勾瑙�")
+model = AutoModel(model="ct-punc")
+res = model.generate(input="閭ｄ粖澶╃殑浼氬氨鍒拌繖閲屽惂 happy new year 鏄庡勾瑙�")
print(res)
```
### Timestamp Prediction
```python
from funasr import AutoModel
-model = AutoModel(model="fa-zh", model_revision="v2.0.0")
-
+model = AutoModel(model="fa-zh")
wav_file = f"{model.model_path}/example/asr_example.wav"
-text_file = f"{model.model_path}/example/asr_example.wav"
-res = model(input=(wav_file, text_file),
- data_type=("sound", "text"))
+text_file = f"{model.model_path}/example/text.txt"
+res = model.generate(input=(wav_file, text_file), data_type=("sound", "text"))
print(res)
```
-[//]: # (FunASR supports inference and fine-tuning of models trained on industrial datasets of tens of thousands of hours. For more details, please refer to ([modelscope_egs](https://alibaba-damo-academy.github.io/FunASR/en/modelscope_pipeline/quick_start.html)). It also supports training and fine-tuning of models on academic standard datasets. For more details, please refer to([egs](https://alibaba-damo-academy.github.io/FunASR/en/academic_recipe/asr_recipe.html)). The models include speech recognition (ASR), speech activity detection (VAD), punctuation recovery, language model, speaker verification, speaker separation, and multi-party conversation speech recognition. For a detailed list of models, please refer to the [Model Zoo](https://github.com/alibaba-damo-academy/FunASR/blob/main/docs/model_zoo/modelscope_models.md):)
+
+More examples ref to [docs](https://github.com/alibaba-damo-academy/FunASR/tree/main/examples/industrial_data_pretraining)
+
+
+## Export ONNX
+
+### Command-line usage
+```shell
+funasr-export ++model=paraformer ++quantize=false ++device=cpu
+```
+
+### Python
+```python
+from funasr import AutoModel
+
+model = AutoModel(model="paraformer", device="cpu")
+
+res = model.export(quantize=False)
+```
+
+### Test ONNX
+```python
+# pip3 install -U funasr-onnx
+from funasr_onnx import Paraformer
+model_dir = "damo/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-pytorch"
+model = Paraformer(model_dir, batch_size=1, quantize=True)
+
+wav_path = ['~/.cache/modelscope/hub/damo/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-pytorch/example/asr_example.wav']
+
+result = model(wav_path)
+print(result)
+```
+
+More examples ref to [demo](runtime/python/onnxruntime)
## Deployment Service
FunASR supports deploying pre-trained or further fine-tuned models for service. Currently, it supports the following types of service deployment:
@@ -209,9 +260,9 @@
You can also scan the following DingTalk group or WeChat group QR code to join the community group for communication and discussion.
-|DingTalk group | WeChat group |
-|:---:|:-----------------------------------------------------:|
-|<div align="left"><img src="docs/images/dingding.jpg" width="250"/> | <img src="docs/images/wechat.png" width="215"/></div> |
+| DingTalk group | WeChat group |
+|:-------------------------------------------------------------------:|:-----------------------------------------------------:|
+| <div align="left"><img src="docs/images/dingding.png" width="250"/> | <img src="docs/images/wechat.png" width="215"/></div> |
## Contributors
@@ -241,10 +292,16 @@
}
@inproceedings{gao22b_interspeech,
author={Zhifu Gao and ShiLiang Zhang and Ian McLoughlin and Zhijie Yan},
- title={{Paraformer: Fast and Accurate Parallel Transformer for Non-autoregressive End-to-End Speech Recognition}},
+ title={Paraformer: Fast and Accurate Parallel Transformer for Non-autoregressive End-to-End Speech Recognition},
year=2022,
booktitle={Proc. Interspeech 2022},
pages={2063--2067},
doi={10.21437/Interspeech.2022-9996}
}
+@inproceedings{shi2023seaco,
+ author={Xian Shi and Yexin Yang and Zerui Li and Yanni Chen and Zhifu Gao and Shiliang Zhang},
+ title={SeACo-Paraformer: A Non-Autoregressive ASR System with Flexible and Effective Hotword Customization Ability},
+ year={2023},
+ booktitle={ICASSP2024}
+}
```
--
Gitblit v1.9.1