Opting In To Improve Speech-to-Text Results
In Audiate, users have the option to opt-in to Deepgram's Model Improvement Partnership Program to help improve Speech-to-Text results.
By default, Audiate users are opted-out of sharing audio data with Deepgram to improve their models. However, users may opt-in to the sharing of this data. This will send anonymous audio to Deepgram for training as you use transcription features in Audiate.
How Do I Opt-in or Opt-out of Deepgram training?
Open Audiate, click "Help" in the menu bar, then click "About"
Here, you can click the checkbox "Improve Speech-to-text with anonymous data" to opt-in (checked) or opt-out (unchecked).
How Is My Information Protected?
Deepgram only stores fractional increments of anonymous collected data, and their infrastructure is compliant with SOC-2 Type 2, GDPR, and other regulatory frameworks.
After opting in, information collected by Deepgram may be retained to train future models. However, they will not redistribute user data to third parties or sell it for marketing purposes.
For more information, please read the article for Deepgram's Model Improvement Partnership Program
https://developers.deepgram.com/docs/the-deepgram-model-improvement-partnership-program