Returning multi-channel recordings as a single file

[Adam]

Hello,

I was researching an issue we just ran into and came across this thread: download Recording with split streams. I'm not sure I fully understand the response from Genesys though.

We tested pulling down recordings via /api/v2/conversations/{ConversationId}/recordings and found two files returned in the response; one for each channel.

If we use the batch recording API, will the response will include a SINGLE file per conversation, even if we have multi-channel enabled on the trunk?

Thanks!

[John_Carnell]

Hi Adam,

I just chatted with the head of our recording team. The two endpoints behave differently by design. The batch recording API will download a single file with both channels combined

The /api/v2/conversations/{ConversationId}/recordings endpoint is meant for integration with Web UIs. It will transcode to whatever codec/formats that are supported by the browser, and it splits it into two separate audio files so that the browser can adjust the volume of the two audio files independently, etc. This /api/v2/conversations/{ConversationId}/recordings endpoint is not the one you should use for speech and text analytics because we do processing on the file specifically to make it consumable by the browser, and it could reduce the fidelity of the audio.

For batch processing (e.g. for speech and text analytics) you should use the batch recording API. This API returns a single file with both channels in it.

I hope that provides some better context.

Thanks,
John Carnell
Manager, Developer Engagement