search

ggerganov / whisper.cpp

Port of OpenAI's Whisper model in C/C++
MIT License
35.71k stars 3.63k forks source link

command example not processing #2390

Open BrianTheBrain11 opened 2 months ago

BrianTheBrain11 commented 2 months ago

I get to "Speech detected! processing..." with no errors. then, no matter what i say, it doesn't seem to "process". Also tried in guided mode. Told me to use one of the commands, same "Speech detected" output but no response when i say any of the commands. At initialization, it picks up my default microphone. When it says "okay whipser, start listening for commands" it will say "speech detected, processing..." no matter what i say. But it wont change no matter what i say, including "ok whisper..." like in the video.

https://github.com/user-attachments/assets/a9064c73-43ff-4b0d-be10-9a9f9b2faf67

I had to clip the video for file restraints, but even leaving it and only it "processing" for a while, on 16 threads, on my ryzen 9, using the small model doens't seem to make any difference. Tried different models as well. After it begins "processing" it just makes my computer cpu usage go way up like its stuck in a loop.

I ran the JFK.wav file example just fine as well. guided mode did not work either. Haven't tried any other example.

BrianTheBrain11 commented 2 months ago

i'm not sure what comment was deleted. if need be i can provide more info.