ggerganov/whisper.cpp

Pull Requests Commits

ggerganov committed 2 years ago

Verified c6174cb8

ggml : sync ggml (clBLAST + tensor names)

ggerganov committed 2 years ago

Verified 0bcb64b1

talk-llama : fix session prompt load (#854)

herrera-luis committed 2 years ago

Verified 0bf680fe

whisper : add detect-language mode (#853)

CRD716 committed 2 years ago

Verified b8064208

talk-llama : add --session support (#845)

herrera-luis committed 2 years ago

Verified be5911a9

bench : improve benchmarks

ggerganov committed 2 years ago

Verified d375d73b

whisper : add memory sizes for Q8_0 (close #846)

ggerganov committed 2 years ago

Verified 7765770f

whisper.wasm : fix typo in readme (#832)

BaffinLee committed 2 years ago

Verified 872a85ae

release : v1.4.1

ggerganov committed 2 years ago

Verified 9c61f5f5

whisper : fix quantize bug (#842)

ggerganov committed 2 years ago

Verified c94c4695

ggml : fix UB (int << 31)

ggerganov committed 2 years ago

Verified feac80dd

release : v1.4.0

ggerganov committed 2 years ago

Verified fa8dbdc8

examples : fix + refactor Levenshtein distance

ggerganov committed 2 years ago

Verified 4a7d49af

whisper : add integer quantization support (#540)

ggerganov committed 2 years ago

Verified 794b162a

whisper : add GPU support via cuBLAS (#834)

ggerganov committed 2 years ago

Verified 5fd1bdd7

ggml : fix WASM build

ggerganov committed 2 years ago

Verified 0ccd6746

ggml : fix 32-bit ARM NEON (#836)

ggerganov committed 2 years ago

Verified d9b550c0

ggml : use vzip instead of vuzp for consistency

ggerganov committed 2 years ago

Verified e9b091c9

ggml : fix WASM build

ggerganov committed 2 years ago

Verified 1f30b992

ggml : sync with ggml repo (warning fixes + asserts)

ggerganov committed 2 years ago

Verified 05c3ea3b

whisper : use correct seek_end when offset is used (#833)

ThijsRay committed 2 years ago

Verified 6108d3cc

tests : add "threads" to run-tests.sh

ggerganov committed 2 years ago

Verified bab97c83

extra : add sync-ggml.sh script

ggerganov committed 2 years ago

Verified 3eaeb030

ggml : sync latest ggml + llama.cpp updates (quantization)

ggerganov committed 2 years ago

Verified acec73ab

whisper.android : add some tips (#816)

Zolliner committed 2 years ago

Verified 5cc17418

build : add WHISPER_COREML_ALLOW_FALLBACK to make / CMake (#812)

ggerganov committed 2 years ago

Verified 3efb81de

whisper : allow non-CoreML fallback when Core ML cannot be loaded (#812)

Canis-UK committed 2 years ago

Verified 94a7cd2a

whisper : fix bug from previous commit

ggerganov committed 2 years ago

Verified 3e82ff47

whisper : avoid designated initializers

ggerganov committed 2 years ago

Verified b5bd2f43

minor : improve C++ and Python style (#768)

asukaminato0721 committed 2 years ago

Verified 94aa56f1

Older