google_cloud_speech_sync_transcribe_local_audio_file.py

1import io
2import os
3
4# Imports the Google Cloud client library
5from google.cloud import speech
6
7
8def run_quickstart() -> speech.RecognizeResponse:
9    # Instantiates a client
10    client = speech.SpeechClient()
11
12    # The name of the audio file to transcribe
13    # Note: This requires a local file named 'audio.raw' to be present.
14    # You can also use a GCS URI like 'gs://cloud-samples-data/speech/brooklyn_bridge.raw'
15    file_name = os.path.join(os.path.dirname(__file__), "resources", "audio.raw")
16
17    # Loads the audio into memory
18    with io.open(file_name, "rb") as audio_file:
19        content = audio_file.read()
20
21    audio = speech.RecognitionAudio(content=content)
22
23    config = speech.RecognitionConfig(
24        encoding=speech.RecognitionConfig.AudioEncoding.LINEAR16,
25        sample_rate_hertz=16000,
26        language_code="en-US",
27    )
28
29    # Detects speech in the audio file
30    response = client.recognize(config=config, audio=audio)
31
32    for result in response.results:
33        print(f"Transcription: {result.alternatives[0].transcript}")
34
35    return response
36
37
38if __name__ == "__main__":
39    run_quickstart()