Customize Your Conversational Intelligence Configuration

Conversational Intelligence supports third-party media recordings. If your call recordings aren't stored in Twilio and you want to use them with Conversational Intelligence, the recordings must be publicly accessible for the duration of transcription. You can host the recordings or use on a time-limited pre-signed URL.

For example, to share a recording on an existing AWS S3 bucket, follow this guide. Then add the public recording URL to the media_url when you create a transcript with the API

If you want to transcribe the audio of a Twilio Video recording, it needs additional processing to create an audio recording that can you can submit for transcription.

To create a dual-channel audio recording first, transcode a separate audio-only composition for each participant in the Video Room.

1
curl -X POST "https://video.twilio.com/v1/Compositions" \ --data-urlencode "AudioSources=PAXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX"
2
\ --data-urlencode "StatusCallback=https://www.example.com/callbacks"
3
\ --data-urlencode "Format=mp4"
4
\ --data-urlencode "RoomSid=RMXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX"
5
\ -u $TWILIO_ACCOUNT_SID:$TWILIO_AUTH_TOKEN

Next, download the media from these compositions and merge them into a single audio stereo audio.

ffmpeg -i speaker1.mp4 -i speaker2.mp4 -filter_complex "[0:a][1:a]amerge=inputs=2[a]" -map "[a]" -f flac -bits_per_raw_smaple 16 -ar 441000 output.flac

In case the recording duration for each participant is different, you can avoid overlapping audio tracks. Use ffmpeg to create a single-stereo audio track with delay to cover the difference in track length. For example, if one audio track last 63 seconds and the other 67 seconds, use ffmpeg to create a stereo file with the first track, with four seconds of delay to match the length of the second track.

ffmpeg -i speaker1.wav -i speaker2.wav -filter_complex "aevalsrc=0:d=${second_to_delay}[s1];[s1][1:a]concat=n=2:v=0:a=1[ac2];[0:a]apad[ac1];[ac1][ac2]amerge=2[a]" -map "[a]" -f flac -bits_per_raw_sample 16 -ar 441000 output.flac

Finally, send a CreateTranscript request to Conversational Intelligence by providing a publicly accessible URL for this audio file as media_url in MediaSource.

By default, Conversational Intelligence assumes Participant one is on channel one, and Participant two is on channel two. If your use case doesn't follow this channel mapping, you can provide optional Participant metadata that maps the participant to the correct audio channel when you create a transcript with the API. You can also use this field to attach other participant metadata to the transcript.

You can provide a CustomerKey when you create a transcript with the API, which allows you to map a Transcript to an internal identifier. This can be a unique identifier within your system to track the transcripts. The CustomerKey is also included as part of the webhook callback when the results for Transcript and Operators become available. This is an optional field, and you can't substitute CustomerKey for Transcript Sid in the APIs.

• Conversational Intelligence enforces rate limits for API resources. You should follow the REST API Best Practices, and implement retries with exponential backoff to properly handle the API response Error 429 "Too Many Requests" for high volume workloads.
• Use Restricted API Keys with Conversational Intelligence for fine-grained control over API resources. For example, you can provide permissions for an API Key that can modify Conversational Intelligence Service resources, but can't access potentially-sensitive information like unredacted Media resources and OperatorResults resources.