def test_sync_recognize(self, mock_create_stub):
# Mock gRPC layer
grpc_stub = mock.Mock()
mock_create_stub.return_value = grpc_stub
client = speech_client.SpeechClient()
# Mock request
encoding = enums.RecognitionConfig.AudioEncoding.FLAC
sample_rate = 44100
config = cloud_speech_pb2.RecognitionConfig(encoding=encoding,
sample_rate=sample_rate)
uri = 'gs://bucket_name/file_name.flac'
audio = cloud_speech_pb2.RecognitionAudio(uri=uri)
# Mock response
expected_response = cloud_speech_pb2.SyncRecognizeResponse()
grpc_stub.SyncRecognize.return_value = expected_response
response = client.sync_recognize(config, audio)
self.assertEqual(expected_response, response)
grpc_stub.SyncRecognize.assert_called_once()
args, kwargs = grpc_stub.SyncRecognize.call_args
self.assertEqual(len(args), 2)
self.assertEqual(len(kwargs), 1)
self.assertIn('metadata', kwargs)
actual_request = args[0]
expected_request = cloud_speech_pb2.SyncRecognizeRequest(config=config,
audio=audio)
self.assertEqual(expected_request, actual_request)
def sync_recognize(self, config, audio, options=None):
"""
Perform synchronous speech-recognition: receive results after all audio
has been sent and processed.
Example:
>>> from google.cloud.gapic.speech.v1beta1 import speech_client
>>> from google.cloud.gapic.speech.v1beta1 import enums
>>> from google.cloud.proto.speech.v1beta1 import cloud_speech_pb2
>>> api = speech_client.SpeechClient()
>>> encoding = enums.RecognitionConfig.AudioEncoding.ENCODING_UNSPECIFIED
>>> sample_rate = 0
>>> config = cloud_speech_pb2.RecognitionConfig(encoding, sample_rate)
>>> uri = ''
>>> audio = cloud_speech_pb2.RecognitionAudio(uri)
>>> response = api.sync_recognize(config, audio)
Args:
config (:class:`google.cloud.proto.speech.v1beta1.cloud_speech_pb2.RecognitionConfig`): [Required] The ``config`` message provides information to the recognizer
that specifies how to process the request.
audio (:class:`google.cloud.proto.speech.v1beta1.cloud_speech_pb2.RecognitionAudio`): [Required] The audio data to be recognized.
options (:class:`google.gax.CallOptions`): Overrides the default
settings for this call, e.g, timeout, retries etc.
Returns:
A :class:`google.cloud.proto.speech.v1beta1.cloud_speech_pb2.SyncRecognizeResponse` instance.
Raises:
:exc:`google.gax.errors.GaxError` if the RPC is aborted.
:exc:`ValueError` if the parameters are invalid.
"""
# Create the request object.
request = cloud_speech_pb2.SyncRecognizeRequest(config=config,
audio=audio)
return self._sync_recognize(request, options)
def main(input_uri, encoding, sample_rate, language_code='en-US'):
service = cloud_speech_pb2.SpeechStub(
make_channel('speech.googleapis.com', 443))
# The method and parameters can be inferred from the proto from which the
# grpc client lib was generated. See:
# https://github.com/googleapis/googleapis/blob/master/google/cloud/speech/v1beta1/cloud_speech.proto
response = service.SyncRecognize(
cloud_speech_pb2.SyncRecognizeRequest(
config=cloud_speech_pb2.RecognitionConfig(
# There are a bunch of config options you can specify. See
# https://goo.gl/KPZn97 for the full list.
encoding=encoding, # one of LINEAR16, FLAC, MULAW, AMR, AMR_WB
sample_rate=sample_rate, # the rate in hertz
# See https://g.co/cloud/speech/docs/languages for a list of
# supported languages.
language_code=language_code, # a BCP-47 language tag
),
audio=cloud_speech_pb2.RecognitionAudio(uri=input_uri, )),
DEADLINE_SECS)
# Print the recognition result alternatives and confidence scores.
for result in response.results:
print('Result:')
for alternative in result.alternatives:
print(u' ({}): {}'.format(alternative.confidence,
alternative.transcript))
def request_transcribe_sync(self, content):
"""
TODO: Description.
:param content:
:return:
"""
service = cloud_speech_pb2.SpeechStub(
self.make_channel('speech.googleapis.com', 443))
# The method and parameters can be inferred from the proto from which the
# grpc client lib was generated. See:
# https://github.com/googleapis/googleapis/blob/master/google/cloud/speech/v1beta1/cloud_speech.proto
response = service.SyncRecognize(
cloud_speech_pb2.SyncRecognizeRequest(
config=cloud_speech_pb2.RecognitionConfig(
# There are a bunch of config options you can specify. See
# https://goo.gl/KPZn97 for the full list.
encoding=self.
encoding, # one of LINEAR16, FLAC, MULAW, AMR, AMR_WB
sample_rate=self.sample_rate, # the rate in hertz
# See https://g.co/cloud/speech/docs/languages for a list of
# supported languages.
language_code=self.language_code, # a BCP-47 language tag
),
audio=cloud_speech_pb2.RecognitionAudio(content=content, )),
DEADLINE_SECS)
# Print the recognition result alternatives and confidence scores.
alternatives = []
for result in response.results:
for alternative in result.alternatives:
alternatives.append({
'service_name': self.config['service_name'],
'confidence': alternative.confidence,
'transcript': alternative.transcript
})
return alternatives