| | |
|---|
| | | We can send streaming audio data to server in real-time with grpc client every 300 ms e.g., and get transcribed text when stop speaking. |
|---|
| | | The audio data is in streaming, the asr inference process is in offline. |
|---|
| | | |
|---|
| | | # Steps |
|---|
| | | |
|---|
| | | ## For the Server |
|---|
| | | |
|---|
| | | Install the modelscope and funasr |
|---|
| | | |
|---|
| | | ```shell |
|---|
| | | pip install "modelscope[audio_asr]" -f https://modelscope.oss-cn-beijing.aliyuncs.com/releases/repo.html |
|---|
| | | pip install -U modelscope funasr |
|---|
| | | # For the users in China, you could install with the command: |
|---|
| | | # pip install -U modelscope funasr -i https://mirror.sjtu.edu.cn/pypi/web/simple |
|---|
| | | git clone https://github.com/alibaba/FunASR.git && cd FunASR |
|---|
| | | pip install --editable ./ |
|---|
| | | ``` |
|---|
| | | |
|---|
| | | Install the requirements for server |
|---|
| | |
|---|
| | | ```shell |
|---|
| | | python ASR_server.py --host "0.0.0.0" --port 10095 --asr_model "damo/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-pytorch" |
|---|
| | | ``` |
|---|
| | | For the paraformer 2pass model |
|---|
| | | |
|---|
| | | ```shell |
|---|
| | | python ASR_server_2pass.py --host "0.0.0.0" --port 10095 --asr_model "damo/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-pytorch" |
|---|
| | | ``` |
|---|
| | | |
|---|
| | | ## For the client |
|---|
| | | |
|---|
| | |
|---|
| | | Start client |
|---|
| | | |
|---|
| | | ```shell |
|---|
| | | python ASR_client.py --host "127.0.0.1" --port 10095 --chunk_size 300 |
|---|
| | | python ASR_client.py --host "127.0.0.1" --port 10095 --chunk_size 50 |
|---|
| | | ``` |
|---|
| | | |
|---|
| | | ## Acknowledge |
|---|
