I am currently reviewing our PCI compliance for our manual call recordings. We are using the ‘Secure Pause’ feature in the agent desktop, but we have found several instances where the recording failed to resume after the agent completed the payment transaction. This has resulted in several lost recordings for our audit trail. Is there a way to automate the ‘Resume’ command via the API so that it is not dependent on the agent manually clicking the button in the UI?
Hello Sho27! I am a speech analytics manager and I have seen these recording gaps impact our sentiment analysis. You can definitely automate the resume command! If your payment application is web-based, you can use the ‘Genesys Cloud Interaction Widget’ to listen for a specific URL or a button click in your app. When the transaction is finished, your app can send a POST request to the /api/v2/conversations/{conversationId}/recordingmetadata endpoint to resume the recording automatically. It is much more reliable than relying on agent memory.
I have been debugging this exact issue for a client. Emm68 is right about the API, but you must ensure that your API client has the conversation:recording:edit permission. Also, keep in mind that there is a slight delay when you send the resume command via the API. If your agent starts talking immediately, you might still lose the first second of the post-payment audio. I recommend adding a small visual indicator in your payment app to show the agent when the recording has successfully resumed.
Greetings. I am also debugging a similar setup. To follow up on Isa47, please check your ‘Session Timeout’ settings in your payment application. If the agent’s payment session times out before they finish the transaction, your automation might never send the resume command, leaving the recording paused indefinitely. We had to add a ‘Fail-Safe’ timer to our middleware that automatically resumes the recording after three minutes if no resume command is received.