eedeebee·2 年前·議論The app has an on device custom audio transcription model that we developed with https://www.stepincto.com for "real time" inferencing of audio input 30 times a second. Happy to discuss more for those interested.