Repository Issue Activity (beta)

#608 Official Jetson/L4T Dockerfile (CUDA on Orin AGX/Nano)

#643 Cant input text into version 0.5.0

#582 Can't generate tts Error!

#961 Reference Text

#960 Kokoro: short prompts (~5-8 words) produce run-on garbage after the text

#723 MLX backend uses CPU variant on Apple Silicon despite gpu_available=true (M5 Pro)

#944 cuda error gtx 1050

#920 Can't download the models in Mac M5

#917 Voicebox is not available in winget

#762 Paste isn't working in v 5.0

#942 Add Qualcomm Hexagon NPU (QNN) backend for Snapdragon X Elite/X Plus

#941 macOS: fn cannot be bound to a capture chord — key_from_str has no "Function" arm, and the resulting Err disables all hotkeys

#934 REST API binds 0.0.0.0 with no auth — destructive endpoints reachable from LAN/VPN

#932 Feature Request: Add Advanced Voice Generation Parameters

#852 Adding voice sample fails: No module named 'audioop' (Python 3.13) — v0.5.0, macOS Apple Silicon

#767 Bug: Linux CUDA backend download installs Windows executable (voicebox-server-cuda.exe)

#841 Offline-guard forces process-global HF_HUB_OFFLINE, blocking lazy per-request downloads (LLM model download + Kokoro preset voices)

#930 Removing the ghost generatations

#803 Tauri panics on Wayland with NVIDIA GPU (tao event loop panic)

#680 [Linux] run `just dev` crashed on startup with `Option::unwrap()` panic in tao event loop

#873 tao panic on Linux set_ignore_cursor_events crashes on transparent windows

#925 Failed model download is reported as "downloading" forever by /models/status, masking real cache state until app restart

#915 Docker build and container startup fail on Windows checkouts due to CRLF line endings

#928 [Bug] Stories tab: voice profile dropdown opens downward below the composer and gets clipped/hidden

#923 voicebox-server memory grows ~unbounded over consecutive generations (CPU backend); once large, new generations hang in "generating" forever