Morph MORPH
SIGN IN SIGN UP
dhruvhead / qvac-ext-lib-whisper.cpp UNCLAIMED
0 0 0 C++

COMMITS

master
July 24, 2026
P
QVAC-21593: tts-cpp: Metal GPU support for Parler-TTS (~2.25x vs CPU, indic q8_0) (#103)
pratiknarola-t committed
67c7ad3
M
QVAC-21928: CosyVoice3 native C++/ggml TTS engine (Qwen2 LM + DiT flow + CausalHiFT) (#99)
Maxym Golovanchuk committed
fc844ce
July 23, 2026
Z
feat(parakeet): add Apple Core ML (Neural Engine) encoder sidecar (#100)
Zbigniew Herman committed
2242355
July 22, 2026
Z
fix(parakeet): long-form windowed encoder on the engines/parakeet layout (#101)
Zbigniew Herman committed
88b690c
F
QVAC-21921 audiogen: native ACE-Step music generation on CPU (#98)
Freddy Martinez Garcia committed
0f832db
O
refactor: land the whisper subtree + umbrella + divergence guard on master (QIP PRs 2+3) (#102)
ogad-tether committed
0c4b3f2
July 21, 2026
O
refactor: move engines to engines/{parakeet,tts} (QIP PR 1) (#95)
ogad-tether committed
07e0983
O
ci: per-engine CI lanes for parakeet-cpp and tts-cpp (QIP PR 0) (#93)
ogad-tether committed
0301210
P
QVAC-19261: tts-cpp: add Parler-TTS engine (mini-v1 / large-v1, CPU) (#92)
pratiknarola-t committed
0775939
July 15, 2026
P
QVAC-21623 whisper: decoder QKV matmul fusion + vocab-logits slice (Adreno q8 decode) (#91)
pratiknarola-t committed
d95e742
July 13, 2026
F
QVAC-21908: chatterbox: make S3Gen cfg_rate configurable (#88)
Freddy Martinez Garcia committed
05879fc
O
feat(parakeet): run the EOU RNN-T decoder as ggml graphs on GPU backends (#85)
ogad-tether committed
ecac5bb
P
feat(lavasr): run denoiser + enhancer on Vulkan (Xclipse 920, 2.76x vs CPU, strict parity) (#89)
pratiknarola-t committed
d2ac458
July 10, 2026
P
feat(lavasr): run the UL-UNAS denoiser on a ggml compute graph (Adreno OpenCL) (#86)
pratiknarola-t committed
2b0fef1
O
parakeet-cpp: repack quantized encoder GEMM weights on CPU (q4_0/q8_0 speedup) (#87)
ogad-tether committed
2e2f4d5
July 7, 2026
M
QVAC-21005 fix: persist all TDT predictor LSTM layer states on GPU (#83)
Maxym Golovanchuk committed
df54e37
Z
QVAC-21726 feat(lavasr): run the enhancer network on a ggml compute graph (GPU + faster CPU) (#82)
Zbigniew Herman committed
1cbea2b
G
QVAC-21550 infra: roll out canonical security baseline (TruffleHog + CodeQL) to qvac-ext-lib-whisper.cpp (#75)
GSServita committed
d16d785
P
tts-cpp: per-op GPU->CPU fallback for TTS-GGML via shared sched_dispatch helper (#81)
pratiknarola-t committed
14ad33b
July 6, 2026
O
QVAC-19557: fix Chatterbox S25/Adreno GPU-load crash — upload quantized weights whole (#79)
ogad-tether committed
1d9ca83
Z
QVAC-21519 chore: remove Asana ticket references from comments and docs (#80)
Zbigniew Herman committed
17af12b
July 3, 2026
Z
QVAC-16579 feat(lavasr): implement UL-UNAS denoiser forward (CPU/GGML) (#78)
Zbigniew Herman committed
6c64f18
I
fix: Add zh to supported languages (#77)
Ishan Vohra committed
9ea1a5e
July 2, 2026
O
tts-cpp: chatterbox-mtl — run quantized KV on the GPU (dequant align-probe cast) (#71)
ogad-tether committed
d149258
Z
QVAC-16579 feat(lavasr): scaffold LavaSR denoiser stage (UL-UNAS) [skeleton] (#76)
Zbigniew Herman committed
032cee1
July 1, 2026
P
QVAC-18192 parakeet-cpp: route compute through ggml_backend_sched (per-op CPU fallback) (#74)
pratiknarola-t committed
7c04e00
June 30, 2026
F
QVAC-21582 — Pull latest from upstream whisper.cpp (v1.9.1) (#73)
Freddy Martinez Garcia committed
cb91a37
Z
QVAC-21483 feat: add output-frequency selection (output_sample_rate) (#69)
Zbigniew Herman committed
ce9ee96
I
QVAC-21335 fix: detect vcpkg mecab headers (#72)
Ishan Vohra committed
28f37ea
June 29, 2026
Z
QVAC-16579 feat: add LavaSR enhancer (CPU/GGML) to tts-cpp (#68)
Zbigniew Herman committed
4c8767a