stiffler01

5 ms·

1. 1.▲
   by stiffler01 3y ago

   That is something we are looking forward to as well. Stay tuned for updates on Jetson support. We tested it on 3090 and 4090 works as expected.
2. 2.▲
   by stiffler01 3y ago

   WhisperLive supports both TensorRT and faster-whisper. We didn’t reduce the chunk size rather use padding based on the chunk size received from the client. Reducing the segment size should be a more optimised solution in the Live scenario.
3. 3.▲
   by stiffler01 3y ago

   Indeed a great point. Waiting for a specific cue, before responding, is an interesting idea. It would make the interaction more natural, especially in situations where the user is thinking aloud or formulating their thoughts before seeking
4. 4.▲
   by stiffler01 3y ago

   We thought about doing this in Whisper itself, since its already working in the audio space.
5. 5.▲
   by stiffler01 3y ago

   Yes, this is something we want to look into in more detail, really appreciate sharing the research.
6. 6.▲
   by stiffler01 3y ago

   Tried this on 4090 and the responsiveness and real-time communication it offers are truly impressive. It has significantly improved my overall experience, especially in scenarios where minimal delay is crucial. Compared to WhisperFusion, Ra