2

I'm using a lambda function to receive a bytes array of audio data, save it as mp3, store it in S3, and then use the S3 object to start a Transcribe job.

Everything's been processed correctly. I can see the .mp3 file in S3. I've also downloaded it to my local machine and played it, and it plays correctly as mp3.

However, when I start the transcription job I get back an error: The media format that you specified doesn't match the detected media format. Check the media format and try your request again.

This is my call to start the AWS Transcribe job:

transcribe.start_transcription_job(
    TranscriptionJobName=job_name,
    Media={'MediaFileUri': job_uri},
    MediaFormat='mp3',
    LanguageCode='en-US'
)

Any idea what may be causing this?

Cheers!

Johan Dingler
  • 61
  • 7

1 Answers1

1

mp3 requires compression, if you just save byte array, then it's not in .mp3 format. You can use soxi to validate audio files: http://sox.sourceforge.net/soxi.html

M--
  • 25,431
  • 8
  • 61
  • 93
Ruoyu Huang
  • 111
  • 4