Ever wondered how listeners would react to your produced audio files from Deepsync? It's kind of a hard thing to find as we like to think that audio quality is purely subjective. Given it's your own voice, we always have a bias for super high quality but our survey shows that consumers are more open to a decent quality that resembles your voice which can lead them to an intent. But how do you know if your script has achieved the quality required for publishing?
Introducing the Listener score.
Deepsync now employs cutting-edge algorithms doing 100s of checks in real-time to predict your listener score. For example - It matches the words spoken by your AI voice to the original script you entered. By comparing various words and phrases, it helps in determining the error rate of your spoken audio. Not only this, it uses AI trained on 1000s of our produced scripts and user feedback in the past to determine what can be improved in your script to increase the listener score.
To try out it for yourself, go to your Studio and create a script. Then, go to your Editor and it will prompt you to find your listener score on the right side. Depending on the length of the script, it might take a few seconds to a few minutes to predict. Please note that this feature is currently in beta and only supports short scripts at this point.
Stay tuned as we ship more updates in order to improve your production experience!