python/FunASR-XL.git

			@@ -23,18 +23,43 @@
			<a name="whats-new"></a>
			## What's new:

			### FunASR runtime-SDK
			### FunASR runtime

			- 2023.07.03:
			We have release the FunASR runtime-SDK-0.1.0, file transcription service (Mandarin) is now supported ([ZH](funasr/runtime/readme_cn.md)/[EN](funasr/runtime/readme.md))

			### Multi-Channel Multi-Party Meeting Transcription 2.0 (M2MeT2.0) Challenge

			We are pleased to announce that the M2MeT2.0 challenge has been accepted by the ASRU 2023 challenge special session. The registration is now open. The baseline system is conducted on FunASR and is provided as a receipe of AliMeeting corpus. For more details you can see the guidence of M2MET2.0 ([CN](https://alibaba-damo-academy.github.io/FunASR/m2met2_cn/index.html)/[EN](https://alibaba-damo-academy.github.io/FunASR/m2met2/index.html)).
			Challenge details ref to ([CN](https://alibaba-damo-academy.github.io/FunASR/m2met2_cn/index.html)/[EN](https://alibaba-damo-academy.github.io/FunASR/m2met2/index.html))

			### Release notes
			### Speech Recognition

			- Academic Models
			- Encoder-Decoder Models (AED): [Transformer](egs/aishell/transformer), [Conformer](egs/aishell/conformer), [Branchformer](egs/aishell/branchformer)
			- Transducer Models (RNNT): [RNNT streaming](egs/aishell/rnnt), [BAT streaming/non-streaming](egs/aishell/bat)
			- Non-autoregressive Model (NAR): [Paraformer](egs/aishell/paraformer)
			- Multi-speaker recognition model: [MFCCA](egs_modelscope/asr/mfcca)

			For the release notes, please ref to [news](https://github.com/alibaba-damo-academy/FunASR/releases)

			- Industrial-level Models
			- Paraformer Models (Mandarin): [Paraformer-large](egs_modelscope/asr/paraformer/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-pytorch), [Paraformer-large-long](egs_modelscope/asr_vad_punc/speech_paraformer-large-vad-punc_asr_nat-zh-cn-16k-common-vocab8404-pytorch), [Paraformer-large streaming](egs_modelscope/asr/paraformer/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-online), [Paraformer-large-contextual](egs_modelscope/asr/paraformer/speech_paraformer-large-contextual_asr_nat-zh-cn-16k-common-vocab8404)
			- Conformer Models (English): [Conformer]()
			- UniASR streaming offline unifying models: [16k UniASR Burmese](https://modelscope.cn/models/damo/speech_UniASR_asr_2pass-my-16k-common-vocab696-pytorch/summary), [16k UniASR Hebrew](https://modelscope.cn/models/damo/speech_UniASR_asr_2pass-he-16k-common-vocab1085-pytorch/summary), [16k UniASR Urdu](https://modelscope.cn/models/damo/speech_UniASR_asr_2pass-ur-16k-common-vocab877-pytorch/summary), [8k UniASR Mandarin financial domain](https://www.modelscope.cn/models/damo/speech_UniASR_asr_2pass-zh-cn-8k-finance-vocab3445-online/summary), [16k UniASR Mandarin audio-visual domain](https://www.modelscope.cn/models/damo/speech_UniASR_asr_2pass-zh-cn-16k-audio_and_video-vocab3445-online/summary),
			[Southern Fujian Dialect model](https://modelscope.cn/models/damo/speech_UniASR_asr_2pass-minnan-16k-common-vocab3825/summary), [French model](https://modelscope.cn/models/damo/speech_UniASR_asr_2pass-fr-16k-common-vocab3472-tensorflow1-online/summary), [German model](https://modelscope.cn/models/damo/speech_UniASR_asr_2pass-de-16k-common-vocab3690-tensorflow1-online/summary), [Vietnamese model](https://modelscope.cn/models/damo/speech_UniASR_asr_2pass-vi-16k-common-vocab1001-pytorch-online/summary), [Persian model](https://modelscope.cn/models/damo/speech_UniASR_asr_2pass-fa-16k-common-vocab1257-pytorch-online/summary)

			- Speaker Recognition
			- Speaker Verification Model: [xvector](egs_modelscope/speaker_verification)
			- Speaker Diarization Model: [SOND](egs/callhome/diarization/sond)

			- Punctuation Restoration
			- Chinese Punctuation Model: [CT-Transformer](egs_modelscope/punctuation/punc_ct-transformer_zh-cn-common-vocab272727-pytorch), [CT-Transformer streaming](egs_modelscope/punctuation/punc_ct-transformer_zh-cn-common-vadrealtime-vocab272727)

			- Endpoint Detection
			- [FSMN-VAD](egs_modelscope/vad/speech_fsmn_vad_zh-cn-16k-common)

			- Timestamp Prediction
			- Character-level FA Model: [TP-Aligner](egs_modelscope/tp/speech_timestamp_prediction-v1-16k-offline)


			<a name="highlights"></a>
			## Highlights
			@@ -75,22 +100,57 @@
			<a name="quick-start"></a>
			## Quick Start

			You could use FunASR by:
			You can use FunASR in the following ways:

			- egs
			- egs_modelscope
			- runtime
			- Service Deployment SDK
			- Industrial model egs
			- Academic model egs

			### egs
			If you want to train the model from scratch, you could use funasr directly by recipe, as the following:
			### Service Deployment SDK

			#### Python version Example
			Supports real-time streaming speech recognition, uses non-streaming models for error correction, and outputs text with punctuation. Currently, only single client is supported. For multi-concurrency, please refer to the C++ version service deployment SDK below.

			##### Server Deployment

			```shell
			cd egs/aishell/paraformer
			. ./run.sh --CUDA_VISIBLE_DEVICES="0,1" --gpu_num=2
			cd funasr/runtime/python/websocket
			python funasr_wss_server.py --port 10095
			```
			More examples could be found in [docs](https://alibaba-damo-academy.github.io/FunASR/en/modelscope_pipeline/quick_start.html)

			### egs_modelscope
			If you want to infer or finetune pretraining models from modelscope, you could use funasr by modelscope pipeline, as the following:
			##### Client Testing

			```shell
			python funasr_wss_client.py --host "127.0.0.1" --port 10095 --mode 2pass --chunk_size "5,10,5"
			```

			For more examples, please refer to [docs](https://alibaba-damo-academy.github.io/FunASR/en/runtime/websocket_python.html#id2).

			#### C++ version Example

			Currently, offline file transcription service (CPU) is supported, and concurrent requests of hundreds of channels are supported.

			##### Server Deployment

			You can use the following command to complete the deployment with one click:

			```shell
			curl -O https://isv-data.oss-cn-hangzhou.aliyuncs.com/ics/MaaS/ASR/shell/funasr-runtime-deploy-offline-cpu-zh.sh
			sudo bash funasr-runtime-deploy-offline-cpu-zh.sh install --workspace ./funasr-runtime-resources
			```

			##### Client Testing

			```shell
			python3 funasr_wss_client.py --host "127.0.0.1" --port 10095 --mode offline --audio_in "../audio/asr_example.wav"
			```

			For more examples, please refer to [docs](https://github.com/alibaba-damo-academy/FunASR/blob/main/funasr/runtime/docs/SDK_tutorial_zh.md)


			### Industrial Model Egs

			If you want to use the pre-trained industrial models in ModelScope for inference or fine-tuning training, you can refer to the following command:

			```python
			from modelscope.pipelines import pipeline
			@@ -105,24 +165,18 @@
			print(rec_result)
			# {'text': '欢迎大家来体验达摩院推出的语音识别模型'}
			```

			More examples could be found in [docs](https://alibaba-damo-academy.github.io/FunASR/en/modelscope_pipeline/quick_start.html)

			### runtime
			### Academic model egs

			An example with websocket:
			If you want to train from scratch, usually for academic models, you can start training and inference with the following command:

			For the server:
			```shell
			cd funasr/runtime/python/websocket
			python funasr_wss_server.py --port 10095
			cd egs/aishell/paraformer
			. ./run.sh --CUDA_VISIBLE_DEVICES="0,1" --gpu_num=2
			```

			For the client:
			```shell
			python funasr_wss_client.py --host "127.0.0.1" --port 10095 --mode 2pass --chunk_size "5,10,5"
			#python funasr_wss_client.py --host "127.0.0.1" --port 10095 --mode 2pass --chunk_size "8,8,4" --audio_in "./data/wav.scp" --output_dir "./results"
			```
			More examples could be found in [docs](https://alibaba-damo-academy.github.io/FunASR/en/runtime/websocket_python.html#id2)
			More examples could be found in [docs](https://alibaba-damo-academy.github.io/FunASR/en/modelscope_pipeline/quick_start.html)

			<a name="contact"></a>
			## Contact