Is there anyway to tweek the app to make it run in real-time (in milliseconds latency) with the Tiny model for example on CPU? My CPU is only 35% when running it, and the latency is more than 10 seconds, on English with Tiny model. Possible to adjust the snippet size (length) or is there any other way to make it real-time? The current speed is not enough as subtitles for meetings and live videos. Thanks!
Is there anyway to tweek the app to make it run in real-time (in milliseconds latency) with the Tiny model for example on CPU? My CPU is only 35% when running it, and the latency is more than 10 seconds, on English with Tiny model. Possible to adjust the snippet size (length) or is there any other way to make it real-time? The current speed is not enough as subtitles for meetings and live videos. Thanks!