Hi, thank you for your awesome job with this project.
I'm trying to use vad with an audio stream, I'm building chunks of 256 samples, and then execute Detect on every chunk.
It is normal that speaking starts in a chunk, and ends in another chunk, but you have a validation that triggers the error "unexpected speech end" if an end is found in a chunk without start.
Hi, thank you for your awesome job with this project. I'm trying to use vad with an audio stream, I'm building chunks of 256 samples, and then execute Detect on every chunk. It is normal that speaking starts in a chunk, and ends in another chunk, but you have a validation that triggers the error "unexpected speech end" if an end is found in a chunk without start.