Speech - RecognitionConfig has no enableAutomaticPunctuation field

Trying to use the Python google speech client lib using the new enableAutomaticPunctuation field but it throws an error. The use_enhanced, model, and metadata fields all work correctly.

So this works:

        config = types.RecognitionConfig(
            encoding=enums.RecognitionConfig.AudioEncoding.FLAC,
            language_code='en-US',
            use_enhanced=True,
            model='phone_call',
            metadata=metadata,
        )

this gives an error:

        config = types.RecognitionConfig(
            encoding=enums.RecognitionConfig.AudioEncoding.FLAC,
            language_code='en-US',
            use_enhanced=True,
            model='phone_call',
            metadata=metadata,
            enableAutomaticPunctuation=True,
        )

  File "/Users/albertlai/venv/lib/python3.5/site-packages/google/protobuf/internal/python_message.py", line 484, in init
    field = _GetFieldByName(message_descriptor, field_name)
  File "/Users/albertlai/venv/lib/python3.5/site-packages/google/protobuf/internal/python_message.py", line 548, in _GetFieldByName
    (message_descriptor.name, field_name))
ValueError: Protocol message RecognitionConfig has no "enableAutomaticPunctuation" field.

Python v3.5.1 google-cloud-speech version v0.33.0 googleapis-common-protos v1.5.3

[The-Gupta]

@nick-jonas, It worked after enabling data logging.

[The-Gupta]

@nick-jonas , even with importing google.cloud.speech_v1p1beta1 I get

ValueError: Protocol message RecognitionConfig has no “model” field.

Same as using google.cloud.speech.