- Newest
- Most votes
- Most comments
There are ways to do this
1 - Use the LEX conversational logs. This will be the transcript.
2 - Use Kinesis Video Stream to fork the audio, and then post process the KVS in lambda to get the WAV file.
Thanks Dan
CX.studio
Here is an example git repo on how to utilize KVS from Amazon Connect to record and analyze the conversation before agent https://github.com/amazon-connect/amazon-connect-realtime-transcription
When using Lex's Conversational Log (transcript and/or Audio to S3), it is only recording what the caller is saying but not how Lex is responding, so you will get multiple wav files per utterance, and then you need to stitch it together.
Here are more reference documentation on KVS with Connect https://docs.aws.amazon.com/connect/latest/adminguide/access-media-stream-data.html
https://docs.aws.amazon.com/connect/latest/adminguide/use-media-streams-blocks.html
For Lex v2 https://docs.aws.amazon.com/lexv2/latest/dg/conversation-logs.html
If you only interested in the Lex portion, I recommend to start with using Conversational Log and all the recent analytics capability of Lex should help you out.
https://docs.aws.amazon.com/lexv2/latest/dg/analytics.html
If you’re imagining a single recording with lex, and maybe DTMF, and then the agent, the answer is no. Perhaps if you forked the call in your SBC before sending it to Connect, but natively what you’re looking for is not possible.
David
Relevant content
- asked 4 months ago
- Accepted Answerasked 2 years ago
- AWS OFFICIALUpdated 2 years ago
- AWS OFFICIALUpdated 3 years ago
- AWS OFFICIALUpdated 2 months ago
- AWS OFFICIALUpdated 3 years ago
Thank you very much for your response. By any chance, do you know how to save the transcript in a S3 bucket using Lex?