0
votes

I'm using python api for Google Cloud Speech to Text. I'm getting the following error when performing client.streaming_recognize

google.api_core.exceptions.InvalidArgument: 400 Unable to recognize speech, code=-73532, possible error in recognition config. Please correct the config and retry the request.

I searched for the error code but didn't find any result.

For the client application I followed this. The web server is Flask based, below is the part where I process data from client.

@socketio.on('initRecording')
def init_recording(data):
    print('Initiated recording on server')

    config = types.RecognitionConfig(
        encoding='LINEAR16',
        sample_rate_hertz=16000,
        language_code='en-US')
    streaming_config = types.StreamingRecognitionConfig(
        config=config, interim_results=False)

    streaming_client = StreamingClient()
    streaming_client.configure(streaming_config)


@socketio.on('binaryData')
def process_recording(data):
    emit('speechData', streaming_client.write(data))


class StreamingClient():
    def __init__(self):
        self.client = speech.SpeechClient()
        self.buffer = []

    def configure(self, streaming_config):
        self.streaming_config = streaming_config

    def write(self, data):
        self.buffer.append(types.StreamingRecognizeRequest(audio_content=data))
        responses = self.client.streaming_recognize(
            self.streaming_config, self.buffer)

        for response in responses:
            if response.error:
                continue

            if not response.results:
                continue

            result = response.results[0]
            if not result.alternatives:
                continue

            if result.is_final:
                self.buffer = []
                return result.alternatives[0].transcript
1
Can you share the code that you are using? is it the sample code of quickstart guide? github.com/GoogleCloudPlatform/python-docs-samples/blob/master/…. Do you have the latest version of the google-cloud-speech library? run pip install --upgrade google-cloud-speech. Take into account that you need to set the GOOGLE_APPLICATION_CREDENTIALS. Follow this documentation: cloud.google.com/speech-to-text/docs/reference/… - Alex Riquelme
@AlexRiquelme added code sample - bit_junky

1 Answers

0
votes

The error code is Google internal, I don't think they are shared. You could create a feature request asking for information in the error message.