add AVX support

Build with OpenBLAS and SDL2 on windows

models : minor changes to the HF convert script (#157)

models : add "convert-h5-to-ggml.py" script (#157)

minor : updates few prints + fix buttons in whisper.wasm

Update README.md

Update README.md

Find libopenblas.dll.a on windows

unicode : fix character replacement (thanks to @tamo)

close #109 : add fetching of the model over HTTP (whisper.wasm)

talk.wasm : final touches

talk.wasm : polishing + adding many AI personalities

stream : "-kc" now enables context keeping from previous segment (#90)

Prompt previous tokens for streaming (#163)

talk.wasm : update README.md

talk.wasm : GPT-2 meets Whisper in WebAssembly (#155)

Update README.md

ggml : fix Windows build

ci : add Windows build

stream : add "max_tokens" cli arg

stream : add "audio_ctx" parameter

stream : add "max_tokens" parameter

stream : add "single_segment" option

stream : partial encoder experiments

fix: free ggml_context (close #149) (#150)

models : simplify the conversion script

Update download-ggml-model.sh

models : change default hosting to Hugging Face

whisper : add option to speed up the audio tempo by x2

make : add libwhisper.so target (#144)

Add WHISPER_NO_AVX and WHISPER_NO_AVX2 to CMakeLists (#136)

minor : remove one more redundant line

minor : fix double float32 conversion in python script

ref #40 : start working on the documentation

Adds support for stdin wav input

js : update whipser.js to latest

Check for AVX and AVX2 on Darwin

Fix the Windows pthread_create shim

sync : submodule whisper.spm

cmake : add submodule whisper.spm

ref #22 : add "duration" option

Update README.md

examples : add simple script for generating Karaoke video

Update README.md

Update README.md

main : fix generated bash script

ggml : multi-thread the ggml_add operator

cmake : fix passing GGML_PERF compile option

Update README.md

whisper : token-level timestamp refactoring (#49, #120)