The real descript-audio-codec package pulls in descript-audiotools,
which transitively requires onnx, tensorboard, protobuf, matplotlib,
pystoi, and other heavy dependencies. onnx fails to build from source
on macOS due to CMake version incompatibility.
TADA only uses Snake1d (a 7-line PyTorch module) from DAC. This commit
adds a shim in backend/utils/dac_shim.py that registers fake dac.*
modules in sys.modules with just the Snake1d class, completely
eliminating the DAC/audiotools dependency chain.
Integrates HumeAI's TADA (Text-Acoustic Dual Alignment) speech-language
model as a new TTS engine. TADA uses a novel 1:1 token-audio alignment
that produces coherent speech over long sequences (700s+).
Two model variants:
- tada-1b: English-only, ~4GB, built on Llama 3.2 1B
- tada-3b-ml: 10 languages, ~8GB, built on Llama 3.2 3B
Backend uses the Encoder for voice prompt encoding with caching, and
TadaForCausalLM with flow-matching diffusion for generation. Supports
bf16 inference on CUDA, forces CPU on macOS (MPS compatibility).
Installed with --no-deps due to torch>=2.7 pin conflict; descript-audio-codec
and torchaudio added as explicit sub-dependencies.
Adds a full effects pipeline powered by Spotify's pedalboard library,
enabling users to apply professional DSP effects (flanger, reverb, delay,
compressor, pitch shift, filters, gain) to generated audio.
Key features:
- Effects chain editor with drag-and-drop reordering (dnd-kit)
- Generation versions: clean copy always saved, processed versions created on top
- Built-in presets (Robotic, Radio, Echo Chamber, Deep Voice) + custom user presets
- Per-profile default effects chain (auto-applied to new generations)
- Per-generation effects override from the generation form
- Apply effects to existing generations from history (creates new version)
- Ephemeral preview endpoint for auditioning effects without persisting
- Dedicated Effects sidebar tab with preset management and live preview
- Version switcher in history cards with expandable panel
- Backward-compatible: existing generations backfilled as clean versions
chatterbox-tts 0.1.6 pins numpy<1.26 and torch==2.6 which are
incompatible with Python 3.12+. Install with --no-deps and list
its sub-dependencies explicitly in requirements.txt.
Also removes HFProgressTracker from chatterbox backend to avoid
'generator didn't stop after throw()' errors from tqdm patching.
- New ChatterboxTTSBackend wrapping ChatterboxMultilingualTTS (ResembleAI/chatterbox)
- Supports 23 languages including Hebrew, forces CPU on macOS (MPS issue)
- Monkey-patches torch.load for CPU loading, forces eager attention for compatibility
- trim_tts_output utility cuts trailing silence/hallucination from Chatterbox output
- Full engine integration: /generate, /generate/stream, model status/download/delete
- Hebrew (he) added to supported languages in frontend and backend validation
- Single flat model dropdown extended with Chatterbox option in both generation UIs
- ModelManagement UI groups LuxTTS and Chatterbox under 'Other Voice Models' section
- Combine engine + model size into one flat dropdown (Qwen3-TTS 1.7B,
Qwen3-TTS 0.6B, LuxTTS) in both FloatingGenerateBox and GenerationForm
- Add linacodec git dep to requirements.txt (uv-only source, pip can't
resolve it from Zipvoice's pyproject.toml)
- Remove redundant transitive deps from requirements.txt
- Quiet the sidecar setup script (was printing misleading instructions)
piper-phonemize has no PyPI wheels — needs custom find-links URL
from k2-fsa.github.io. Removed redundant transitive deps that
Zipvoice already declares.
Introduce LuxTTS (ZipVoice) alongside Qwen TTS, enabling users to choose
between engines at generation time. LuxTTS offers fast, English-focused
voice cloning at 48kHz with ~1GB VRAM.
Backend:
- Add LuxTTSBackend with encode_prompt/generate_speech integration
- Multi-engine registry (get_tts_backend_for_engine) replacing singleton
- Engine-prefixed voice prompt cache keys to avoid collisions
- Engine field on GenerationRequest (default 'qwen' for backward compat)
- Engine dispatch in /generate and /generate/stream endpoints
- LuxTTS in model status, download, and delete maps
Frontend:
- TTS Engine selector dropdown in GenerationForm (Qwen TTS / LuxTTS)
- Conditionally hide Model Size and Delivery Instructions for LuxTTS
- Engine field added to TypeScript types and Zod schema
- LuxTTS section in Model Management page
Add the ability to download a CUDA-enabled backend binary (~2.4 GB) and
swap it in via a backend-only restart, solving the #1 user pain point
(19 open 'GPU not detected' issues caused by GitHub's 2 GB asset limit).
Backend:
- cuda_download.py: download from R2 (primary) or GitHub split-parts
(fallback), SHA-256 verification, atomic writes, progress via SSE
- 4 new endpoints: GET/POST/DELETE /backend/cuda-*, GET cuda-progress
- server.py: --version flag, auto-detect variant from binary name
- build_binary.py: --cuda flag for CUDA PyInstaller builds
- split_binary.py: split large binaries into <2GB GitHub Release assets
- CI workflow for building CUDA binary
Tauri:
- restart_server command (stop -> wait -> start)
- start_server prefers CUDA binary from {data_dir}/backends/ if present
- Version mismatch check: runs --version before launching CUDA binary
Frontend:
- GpuAcceleration component: download, progress, restart, switch, delete
- API client + types for CUDA status and management
- Platform lifecycle: restartServer() on Tauri/Web
- Aggressive 1s health polling during restart for fast reconnection
- Added functionality to upload, delete, and retrieve avatar images for voice profiles.
- Introduced new API endpoints for avatar management, including upload and delete operations.
- Enhanced profile forms and components to support avatar image handling, including previews and error handling.
- Updated database schema to include avatar_path for profiles and added necessary migrations.
- Implemented image validation and processing utilities to ensure proper avatar uploads.
James Pine