PIT Transcription Tool

pit_transcriptor.py

from google.cloud import speech
from google.cloud.speech import enums
from google.cloud.speech import types

from google.cloud.storage import Blog

import sys

speech_client = speech.SpeechClient()
base_uri = sys.argv[1]

audio = types.RecognitionAudio(uri=base_uri)
config = types.RecognitionConfig(
        encoding=enums.RecognitionConfig.AudioEncoding.FLAC,
        language_code='en-GB')

operation = speech_client.long_running_recognize(config, audio)

audio_lines = []
response = operation.result()

for result in response.results:
    script = result.alternatives
    confidence = script[0].confidence

    if len(script) > 1:
        alternatives = '; '.join(result.alternatives[1:])
    else:
        alternatives = 'No Alternatives'

    message = f'{script[0]} :Confidence:{confidence} Alternatives: {alternatives}'
    audio_lines.append(message)

transcript = '\n'.join(audio_lines)
with open(f'transcript_{base_uri}', 'w') as transcript_file:
    transcript_file.write(transcript)