Skip to content

v0.2.0

Latest
Latest

Choose a tag to compare

View all tags
@github-actions github-actions released this 02 Oct 15:42
· 9 commits to main since this release
v0.2.0
This tag was signed with the committer’s verified signature.
pskrunner14 Prabhsimran Singh
GPG key ID: BC440D848CE64065
Verified
Learn about vigilant mode.
6a3ca36
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.

Highlights

  • Nemotron 3 Diarization. Support for the high-resolution Sortformer architecture with native 10 ms output, for both standalone diarization and diarized ASR. (#50, #52)
  • MagpieTTS v2607. Frame-stacking support, with Arabic, Korean and Portuguese added. The model index now points to the v2607 GGUF. (#17, #42)
  • Faster MagpieTTS on CUDA and CPU. Persistent decoder kernels, fused NanoCodec convolutions, prefetching across text chunks, and faster CPU GEMMs. See BENCHMARK.md for measured results. (#51)
  • Nemotron Labs VoiceChat. Full-duplex speech-to-speech with realtime audio, transcripts and tool calling, served over the realtime WebSocket API. VoiceChat is CUDA-only and is built from source with the cuda-s2s preset; it isn't included in the release archives. See the VoiceChat guide. (#21, #39)
  • Live diarized transcription. transcribe --live --diarize shows speaker turns as you speak, and --live now enables endpointing by default. (#50)
  • Upgraded llama.cpp. ggml now comes from a single pinned llama.cpp submodule (b11151) instead of a separate ggml submodule. The project's patches in patches/ are applied automatically at configure time. (#57)

Fixes

  • MagpieTTS no longer returns silence on M5-class Macs. (#20)
  • Fixed nondeterministic NanoCodec audio corruption on CUDA, caused by a fused kernel reading a buffer the graph allocator had reused. (#28)
  • MagpieTTS builds and runs on stock ggml (Metal, Vulkan) without the patched fused attention. (#24, #51)
  • Streaming ASR keeps acoustic and decoder context across endpoints, and late punctuation no longer leaks into the next final. (#50)
  • Skinny-Q8 CUDA results no longer depend on earlier requests. (#51)
  • Linux Vulkan archives use the system C++ runtime, so Mesa drivers (AMD, Intel) load on current distributions. (#58)

Notes

  • Nightly builds. Install the latest main with --channel nightly, rebuilt daily. (#58)
  • Text normalization. Linux and macOS archives now include inverse text normalization (ITN) and text normalization (TN). (#58)

New Contributors

Full Changelog: v0.1.0...v0.2.0

Contributors

danilogr, ryanleary, and virajkarandikar
Assets 35
Loading