offline-voice

85 / 100 · Low risk

Fine to try.

By notivn/AIEVv2 (GitHub) · media · 142 stars · scanned at commit b11efbd

How the on-device narration engine (VieNeu-TTS) and voice cloning work in this system - install tiers, the worker protocol, and the verified traps around torchaudio, voice metadata parsing and duration. Read when working on Text to video narration, the Voices page, the /api/voices or /api/tts endpoints, or when a user reports that offline voices or voice cloning fail to install or sound wrong.

Works with any agent that supports SKILL.md skills.

What this skill can do

Downloads and installs other software

The author does not mention this.

Found in 6 places.

.claude/skills/offline-voice/SKILL.md, line 18| Setup | `GEMINI_API_KEY` | `pip install vieneu` |
.claude/skills/offline-voice/SKILL.md, line 401. **Speech only**: `pip install vieneu`. Torch-free (ONNX Runtime), ~30 MB of wheels. This is enough for all 14 preset voices.
.claude/skills/offline-voice/SKILL.md, line 412. **Cloning**: additionally `pip install torch torchaudio`, 2-3 GB.

Show 3 more

.claude/skills/offline-voice/SKILL.md, line 50**Cause:** PyAV 17+ dropped `metadata_errors`, but faster-whisper 1.2.x still passes it to `av.open()` (verified 2026-10: `av` 19.0.1 + `faster-whisper` 1.2.1).
.claude/skills/offline-voice/SKILL.md, line 51**Fix:** `.runtime/venv/Scripts/python.exe -m pip install "av>=11,<17"` (16.1.0 verified). `start/doctor.mjs` now pins it in the faster-whisper fix, but a venv
.claude/skills/offline-voice/SKILL.md, line 75**Fix:** never call either. The store lives at `assets/voices/` (`library.json` plus `<id>/ref.wav`) and voices are re-registered into the worker on demand with

Install

Paste this into your terminal. It downloads exactly the version we scanned, not whatever is newest.

if [ -e ~/.agents/skills/'offline-voice' ] || [ -L ~/.agents/skills/'offline-voice' ]; then echo 'A skill folder named offline-voice already exists — not replacing it' >&2; false; else d=$(mktemp -d) && git clone --quiet --filter=blob:none --no-checkout 'https://github.com/notivn/AIEVv2' "$d" && git -C "$d" checkout --quiet 'b11efbdf9ce007100701eca69ae7127e7da51e56' -- '.claude/skills/offline-voice' && mkdir -p ~/.agents/skills && cp -R "$d"/'.claude/skills/offline-voice' ~/.agents/skills/'offline-voice' && rm -rf "$d"; fi

Codex reads ~/.agents/skills directly. For Claude Code, Cursor, Copilot, Gemini CLI, Antigravity or OpenCode, also link it. For Claude Code: ln -s ~/.agents/skills/'offline-voice' ~/.claude/skills/'offline-voice' (other agents: ~/.cursor/skills, ~/.copilot/skills, ~/.gemini/skills, ~/.gemini/config/skills, ~/.config/opencode/skills). Or install through the finder skill, which does both.

Badge for your README

The badge for notivn/AIEVv2 shows the score of its lowest-scoring skill.

SkillTrust badge for notivn/AIEVv2

[![SkillTrust](https://zalavijaysinh.github.io/skilltrust/badges/notivn/AIEVv2.svg)](https://zalavijaysinh.github.io/skilltrust/repos/notivn/AIEVv2/)

Score history

Date	Version	Score
2026-11-07	`b11efbd`	85