def main(): args = BaseRecognitionParser().parse_args() if args.encoding == stt_pb2.RAW_OPUS: raise ValueError("RAW_OPUS encoding is not supported by this script") with audio_open_read(args.audio_file, args.encoding, args.rate, args.num_channels, args.chunk_size, args.pyaudio_max_seconds) as reader: metadata = authorization_metadata(args.api_key, args.secret_key, "tinkoff.cloud.stt", type=dict) request = build_recognition_request(args, reader, type="json") response = requests.post("http{}://{}/v1/stt:recognize".format( "s" if args.endpoint.endswith("443") else "", args.endpoint), json=request, headers=metadata) if response.status_code != 200: print( "REST failed with HTTP code {}\nHeaders: {}\nBody: {}".format( response.status_code, response.headers, response.text)) return response = response.json() print_recognition_response(response)
def main(): args = StreamingRecognitionParser().parse_args() with audio_open_read(args.audio_file, args.encoding, args.rate, args.num_channels, args.chunk_size, args.pyaudio_max_seconds) as reader: stub = stt_pb2_grpc.SpeechToTextStub(make_channel(args)) metadata = authorization_metadata(args.api_key, args.secret_key, "tinkoff.cloud.stt") responses = stub.StreamingRecognize(stt_generate_requests(args, reader), metadata=metadata) print_streaming_recognition_responses(responses)
def main(): args = BaseRecognitionParser().parse_args() if args.encoding == stt_pb2.RAW_OPUS: raise ValueError("RAW_OPUS encoding is not supported by this script") with audio_open_read(args.audio_file, args.encoding, args.rate, args.num_channels, args.chunk_size, args.pyaudio_max_seconds) as reader: stub = stt_pb2_grpc.SpeechToTextStub(make_channel(args)) metadata = authorization_metadata(args.api_key, args.secret_key, "tinkoff.cloud.stt") response = stub.Recognize(build_recognition_request(args, reader), metadata=metadata) print_recognition_response(response)
def main(): args = BaseRecognitionParser().parse_args() total = '' if args.encoding == stt_pb2.RAW_OPUS: raise ValueError("RAW_OPUS encoding is not supported by this script") with audio_open_read(args.audio_file, args.encoding, args.rate, args.num_channels, args.chunk_size, args.pyaudio_max_seconds) as reader: stub = stt_pb2_grpc.SpeechToTextStub(make_channel(args)) metadata = authorization_metadata(args.api_key, args.secret_key, "tinkoff.cloud.stt") response = stub.Recognize(build_recognition_request(args, reader), metadata=metadata) if not isinstance(response, dict): # https://developers.google.com/protocol-buffers/docs/proto3#json response = MessageToDict(response, including_default_value_fields=True, preserving_proto_field_name=True) for result in response["results"]: for alternative in result["alternatives"]: total = total + alternative["transcript"] print(total)