asr-client-code

This repo including sample clients to connect with NTU AISpeechLab server The audio format should be either in .wav, .mp3 or raw.

Online scripts python <script> -m -u <server_url> -r -t <your_token> <path_to_audio_file>

Where
<mode>: must be 'file' or 'stream'
<path_to_audio_file>: is required in file mode, but for stream it isn't required
<your_token>: Please get the token from AISpeechLab
<user_url>: ws://HOST:PORT/client/ws/speech

• client.py

  • Connect with AISpeechLab online ASR via websocket, simulate streaming audio from file
  • Example usage: python client.py -u <server_url> -r -t <your_token> <path_to_audio_file>

• client2.py

  • Connect with AISpeechLab online ASR via websocket, simulate streaming audio from file or microphone, using Python2.
  • Example usage: python client2.py -m -u <server_url> -r -t <your_token> <path_to_audio_file>

• client3.py

  • Connect with AISpeechLab online ASR via websocket, simulate streaming audio from file or microphone, using Python3.
  • Example usage: python client3.py -m -u <server_url> -r -t <your_token> <path_to_audio_file>

• offline_client.py

  • Connect with AISpeechLab offline ASR via HTTP POST request, including 3 steps
    • Upload your audio file -> get the speechId
    • Check the processing status based on returned speechId
    • Download the transcription when the processing is done
  • Example usage