使用Google Cloud Speech转录音频文件时如何解决“请求包含无效的参数错误”

Question

We are able to create unique buckets on each run of the program, however, it hits a road-block around the time it reaches the transcribe_gcs function. We want the program to transcribe the audio file that gets uploaded to the bucket. But the transcribe process isn't quite functioning.

We changed the directory for the gcs_uri to be "gs://". This allows unique buckets to be created each time.

def transcribe_gcs(gcs_uri):
    """Asynchronously transcribes the audio file specified by the gcs_uri."""
    #from google.cloud import speech
    from google.cloud.speech import enums
    from google.cloud.speech import types
    from google.cloud.speech_v1p1beta1 import enums
    from google.cloud.speech_v1p1beta1 import types
    audio = types.RecognitionAudio(uri=gcs_uri)
    config = types.RecognitionConfig(
        encoding='LINEAR16',
        sample_rate_hertz=44100,
        language_code='en-US',
        enable_speaker_diarization=True,
        diarization_speaker_count=2)

    client = speech.SpeechClient()

    ##response = client.recognize(config, audio)

    operation = client.long_running_recognize(config, audio)
    print('Waiting for operation to complete...')
    response = operation.result(timeout=3000)
    result = response.results[-1]

    words_info = result.alternatives[0].words

    tag = 1
    speaker = ""

    for word_info in words_info:
        if word_info.speaker_tag == tag:
            speaker = speaker + " " + word_info.word     #need to adjust how speakers are actually separated

        else:
            print("Speaker {}: {}".format(tag, speaker)) #get program to print entire transcript through here
            tag = word_info.speaker_tag
            speaker = "" + word_info.word                #make sentiment analysis work on each individual line


    # Each result is for a consecutive portion of the audio. Iterate through
    # them to get the transcripts for the entire audio file.
    for result in response.results:
        # The first alternative is the most likely one for this portion.
        print(u'Transcript: {}'.format(result.alternatives[0].transcript)) #this should be removed eventually but should be used somehow to modify the speaker portion

        transcribedSpeechFile = open('speechToAnalyze.txt', 'a+')  # this is where a text file is made with the transcribed speech

        transcribedSpeechFile.write(format(result.alternatives[0].transcript))

        transcribedSpeechFile.close()

        confidencePercentage = result.alternatives[0].confidence
        confidencePercentage = confidencePercentage * 100

        print("Confidence level of transcription: {}%".format(round(confidencePercentage, 2)))
# [END speech_transcribe_async_gcs]


if __name__ == '__main__':
    transcribe_gcs(gcs_uri)

Expected results: transcribes the audio file uploaded to the unique bucket

Actual results: creates a bucket, but does not go further than that.

ERROR:

Traceback (most recent call last):
  File "C:\Users\Dave\AppData\Roaming\Python\Python37\site-packages\google\api_core\grpc_helpers.py", line 57, in error_remapped_callable
    return callable_(*args, **kwargs)
  File "C:\Users\Dave\AppData\Roaming\Python\Python37\site-packages\grpc\_channel.py", line 565, in __call__
    return _end_unary_response_blocking(state, call, False, None)
  File "C:\Users\Dave\AppData\Roaming\Python\Python37\site-packages\grpc\_channel.py", line 467, in _end_unary_response_blocking
    raise _Rendezvous(state, None, None, deadline)
grpc._channel._Rendezvous: <_Rendezvous of RPC that terminated with:
    status = StatusCode.INVALID_ARGUMENT
    details = "Request contains an invalid argument."
    debug_error_string = "{"created":"@1564207941.288000000","description":"Error received from peer ipv6:[2607:f8b0:4000:80e::200a]:443","file":"src/core/lib/surface/call.cc","file_line":1052,"grpc_message":"Request contains an invalid argument.","grpc_status":3}"
>

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
  File "C:/Users/Dave/Desktop/mizu/test.py", line 120, in <module>
    transcribe_gcs(gcs_uri)
  File "C:/Users/Dave/Desktop/mizu/test.py", line 80, in transcribe_gcs
    operation = client.long_running_recognize(config, audio)
  File "C:\Users\Dave\AppData\Local\Programs\Python\Python37\lib\site-packages\google\cloud\speech_v1p1beta1\gapic\speech_client.py", line 326, in long_running_recognize
    request, retry=retry, timeout=timeout, metadata=metadata
  File "C:\Users\Dave\AppData\Roaming\Python\Python37\site-packages\google\api_core\gapic_v1\method.py", line 143, in __call__
    return wrapped_func(*args, **kwargs)
  File "C:\Users\Dave\AppData\Roaming\Python\Python37\site-packages\google\api_core\retry.py", line 273, in retry_wrapped_func
    on_error=on_error,
  File "C:\Users\Dave\AppData\Roaming\Python\Python37\site-packages\google\api_core\retry.py", line 182, in retry_target
    return target()
  File "C:\Users\Dave\AppData\Roaming\Python\Python37\site-packages\google\api_core\timeout.py", line 214, in func_with_timeout
    return func(*args, **kwargs)
  File "C:\Users\Dave\AppData\Roaming\Python\Python37\site-packages\google\api_core\grpc_helpers.py", line 59, in error_remapped_callable
    six.raise_from(exceptions.from_grpc_error(exc), exc)
  File "<string>", line 3, in raise_from
google.api_core.exceptions.InvalidArgument: 400 Request contains an invalid argument

Answer 1

Have you tried with Google Cloud's uri sample yet?

gs://gcs-test-data/vr.flac

Answer 2

After making some changes to your code as @siamsot suggested in his comment I could reproduce the error that you got. It occurs only when you don't pass a valid gcs_uri .

It should be of type string and with the format:

gs://[BUCKET_NAME]/[PATH_TO_FILE]/[FILENAME]

like the google sample that @Huy Nguyen posted in their answer:

gs://gcs-test-data/vr.flac

I suspect that you didn't specify the filename or the prefix gs:// in the gcs_uri . I managed to transcribe the above sample file with your code. If you want to test it, change your imports to:

from google.cloud import speechv1p1beta1 as speech
#from google.cloud.speech import enums
#from google.cloud.speech import types
#from google.cloud.speech_v1p1beta1 import enums
from google.cloud.speech_v1p1beta1 import types

and pass 'gs://gcs-test-data/vr.flac' as the gcs_uri to the transcribe_gcs function.

As this file is different from what you are expecting in your code you should change the encoding and sample_rate_hertz properties of RecognitionConfig to 'FLAC' and 16000 respectively.