Issue when Transcription Enabled

[rajrishi7817]

I see error when I enable transcription. CloudWatch logs says that "Your request timed out because no new audio was received for 15 seconds." I have configured START_SELECTOR_TYPE as FRAGMENT_NUMBER. I do have Lex inputs in Contact Flow. I am sure audio is provided both from IVR and caller within 15s of lambda trigger.

Note: When transcription is not enabled, the function stores audio files in S3 without any issues/error.

Could you assist with the reason of this issue.

[angieyu]

Hi is this still an issue? Did you manage to fix it?