faster whisper infer - Githubissues

k2-fsa / sherpa-onnx

Speech-to-text, text-to-speech, speaker diarization, and VAD using next-gen Kaldi with onnxruntime without Internet connection. Support embedded systems, Android, iOS, Raspberry Pi, RISC-V, x86_64 servers, websocket server/client, C/C++, Python, Kotlin, C#, Go, NodeJS, Java, Swift, Dart, JavaScript, Flutter, Object Pascal, Lazarus, Rust

https://k2-fsa.github.io/sherpa/onnx/index.html

Apache License 2.0

3.69k stars 428 forks source link

faster whisper infer #1559

Open didadida-r opened 4 days ago

didadida-r commented 4 days ago

hi, i want to know where i can use a faster whisper infer scirpt after seeing the PR faster whisper llm trt. i only want to transcribe using the whisper model, not whisper llm in sherpa triton. could you please direct me to the official script or documentation for the optimal and most accelerated version of the whisper model?

the log in whisper in Run with GPU (int8) doc is

decoding method: greedy_search
Elapsed seconds: 19.190 s
Real time factor (RTF): 19.190 / 6.625 = 2.897