You want to process audio file into text.
\nWe will use gapi which offers ability to transform audio files into text.
\n# GOOGLE_APPLICATION_CREDENTIALS=./gapi-auth.json\nimport json\nimport subprocess\n\nfrom google.cloud import speech\nfrom google.cloud.proto.speech.v1 import cloud_speech_pb2\n\nspeech_client = speech.SpeechClient()\n\n\npath_to_file = './input.mp4'\noutput_path_to_file = './out.flac'\n\n\n# command = 'ffmpeg -i {input_file} -ac 1 -c:a flac answer_11.flac'\nsubprocess.call(['ffmpeg', '-i', '{input_file}'.format(input_file=path_to_file), '-ac', '1', '-c:a', 'flac', output_path_to_file])\n\n\ndef parse_speech_recognition_result(speech_recognition_result):\n data = {'data': []}\n for result in speech_recognition_result.results:\n for alternative in result.alternatives:\n data['data'].append({\n 'transcript': alternative.transcript,\n 'confidence': alternative.confidence,\n })\n return data\n\n\nwith open(output_path_to_file, 'rb') as recording_file:\n recording_bytes = recording_file.read()\n audio = cloud_speech_pb2.RecognitionAudio(content=recording_bytes)\n config = cloud_speech_pb2.RecognitionConfig(encoding=\"FLAC\", sample_rate_hertz=44100, language_code=\"en-US\")\n speech_recognition_result = speech_client.recognize(config=config, audio=audio)\n result = parse_speech_recognition_result(speech_recognition_result)\n\n dump_result = json.dumps(result)\n\n print(dump_result)