Skip to content

Actions: mybigday/codec.cpp

Actions

CI

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
ci.yml
67 workflow runs
67 workflow runs

Filter by Event

⌥ + click/return to exclude

Filter by Status

⌥ + click/return to exclude

Filter by Branch

⌥ + click/return to exclude

Filter by Actor

⌥ + click/return to exclude
fix(chatterbox-s3g): stop double-applying the iSTFT synthesis window CI #67: Commit f8ceb01 pushed by hans00
7m 32s main
main
7m 32s
fix(xy-tokenizer): chunked decode for long sequences (pos_emb table i… CI #66: Commit c5ef02b pushed by hans00
7m 32s main
main
7m 32s
feat(tts): use llama.cpp common_sampler for backbone logits + prompt-… CI #65: Commit cc9e4cb pushed by hans00
6m 57s main
main
6m 57s
fix(bluemagpie): sample N(0,1) CFM init noise when caller passes NULL CI #64: Commit e0f8343 pushed by hans00
1m 37s main
main
1m 37s
perf(bluemagpie): fused SwiGLU + hoisted rope positions + in-graph KV… CI #63: Commit 17714fb pushed by hans00
1m 39s main
main
1m 39s
perf(bluemagpie): drop per-eval weight dequant + fuse QKV/gate-up + n… CI #62: Commit cf61db7 pushed by hans00
1m 35s main
main
1m 35s
chore: trim CODEC_BUILD_TTS_CLI option help text CI #61: Commit d0c825f pushed by hans00
1m 32s main
main
1m 32s
build(barbet): apply llama.cpp patch via CMake instead of a manual step CI #60: Commit 22ea422 pushed by hans00
1m 45s main
main
1m 45s
chore: drop stale submodule comment from .gitignore CI #59: Commit bd31cce pushed by hans00
1m 43s main
main
1m 43s
build(barbet): pin llama.cpp as submodule under common/third-party CI #58: Commit 57a8354 pushed by hans00
1m 37s main
main
1m 37s
feat(bluemagpie): BlueMagpie-TTS continuous-latent CFM codec + codec_… CI #57: Commit 11b75fb pushed by hans00
1m 42s main
main
1m 42s
feat(common): codec_common step 5 — observe_codes for Type C / Type D CI #56: Commit 2d3cd80 pushed by hans00
1m 48s main
main
1m 48s
test(moss-ttsd): codec.cpp encoder parity vs HF processor CI #55: Commit 51b9338 pushed by hans00
1m 38s main
main
1m 38s
feat(tts): chatterbox T3 native — Llama 520M backbone + S3G codec CI #54: Commit 5a78e8a pushed by hans00
1m 39s main
main
1m 39s
docs(audio): MTMD streaming verdict + codec_lm completion plan CI #53: Commit c7e3f4e pushed by hans00
1m 41s main
main
1m 41s
docs: mark moss-tts-nano-100m as shipped (HF AR + codec.cpp decode) CI #52: Commit a468467 pushed by hans00
2m 9s main
main
2m 9s
cleanup: drop per-model eval-arena hint in favor of fixed default CI #51: Commit 180f34e pushed by hans00
1m 49s main
main
1m 49s
perf: drop name-based output flag heuristic; mark snac codes explicitly CI #50: Commit ef658a8 pushed by hans00
1m 37s main
main
1m 37s
perf: switch graph alloc to ggml_backend_sched_alloc_graph CI #49: Commit 1e4ae21 pushed by hans00
1m 43s main
main
1m 43s
refactor: replace all ggml_map_custom* ops with native ggml graphs CI #48: Commit a6c0a16 pushed by hans00
1m 42s main
main
1m 42s
fix(e2e/runner): hoist importlib.util import to module level CI #47: Commit cd7b6c7 pushed by hans00
2m 57s main
main
2m 57s
refactor: extract Whisper layer, mel-fbank, L2-norm, conv1d squeeze CI #46: Commit bf6c92d pushed by hans00
1m 41s main
main
1m 41s
feat: add OpenMOSS XY-Tokenizer (16 kHz → 24 kHz neural codec) CI #45: Commit 38b352e pushed by hans00
1m 39s main
main
1m 39s
feat: support full MOSS-Audio-Tokenizer (1.6B mono 24 kHz) CI #44: Commit 7d68473 pushed by hans00
1m 36s main
main
1m 36s
feat: add HKUSTAudio/xcodec2 with full encode + decode parity CI #43: Commit 8bb791e pushed by hans00
1m 52s main
main
1m 52s