Script que obtiene la transcripción de un audio alojado en Cloud Storage

audio-transcription.py

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

AUDIO_CHANNELS = 2
AUDIO_RATE = 16000
BLOB_PATH = "audio.flac"
BUCKET_NAME = "audio-devlife"
LANG_CODE = 'es-MX'


with audioread.audio_open(BLOB_PATH) as f:
    AUDIO_CHANNELS = int(f.channels)
    AUDIO_RATE = int(f.samplerate)

gcs_uri = "gs://{}/{}".format(BUCKET_NAME, BLOB_PATH)
client = speech.SpeechClient()

audio = types.RecognitionAudio(uri=gcs_uri)

config = types.RecognitionConfig(
    encoding=enums.RecognitionConfig.AudioEncoding.FLAC,
    language_code=LANG_CODE,
    audio_channel_count=AUDIO_CHANNELS,
    enable_separate_recognition_per_channel=True)

operation = client.long_running_recognize(config, audio)
response = operation.result()

for result in response.results:
    for alternative in result.alternatives:
        print('=' * 20)
        print(alternative.transcript)
        print(alternative.confidence)