get_transcription_text.py

def get_transcription_text(transcribe, job_name):
    """Returns transcription text for the AWS Transcribe job
    
    Parameters
    ----------
    transcribe : AWS `transcribe` service client instance
    job_name : transcribe service `job` name
    
    Returns
    -------
    Current job status if job is still in progress
    Transcription text if job is Completed
    """
    import urllib.request
    import json
    import time
    
    # let's obtain the job instance
    job = transcribe.get_transcription_job(TranscriptionJobName=job_name)
    
    # and it's status
    status = job['TranscriptionJob']['TranscriptionJobStatus']
    
    # check the status every 5 seconds and 
    # return the transcribed text if the job is finished
    # otherwise return None if job failed
    while True:
        if status == 'COMPLETED':
            print(f"Job {job_name} completed")
            with urllib.request.urlopen(job['TranscriptionJob']['Transcript']['TranscriptFileUri']) as r:
                data = json.loads(r.read())
            return data['results']['transcripts'][0]['transcript']
        elif status == 'FAILED':
            print(f"Job {job_name} failed")
            return None
        else:
            print(f"Status of job {job_name}: {status}")
            time.sleep(5)