Use this skill when the user asks for realistic, expressive speech workflows that should be ready for delivery in real scenarios.
Repository
Capabilities
- Generate natural-sounding TTS with speaking style presets
- Tune fillers, pacing, and emotional tone for companion-like output
- Run local-first or cloud-backed workflows depending on privacy/speed needs
- Produce delivery-ready audio for voice broadcast and chat app workflows
Recommended flow
- Clarify use case and audience (news, podcast, assistant voice, alert).
- Pick backend mode:
- Local-first for privacy-sensitive drafts
- Cloud backend for speed and expressive controls
- Generate short validation samples before long renders.
- Apply style tuning (emotion/fillers/pacing) only where it improves clarity.
- Render final audio and verify duration, clipping, and output format.
- Package output for downstream app delivery.
Quick commands
# List available NoizAI skills
npx skills add NoizAI/skills --list --full-depth
# Install TTS skill
npx skills add NoizAI/skills --full-depth --skill tts -y
# Example verification command
bash skills/characteristic-voice/scripts/speak.sh \
--preset comfort -t "Hmm... I'm right here." -o comfort.wav
Guardrails
- Keep descriptions factual and avoid overclaiming quality.
- If a model/backend is unavailable, provide fallback options.
- Do not expose API keys in logs, commits, or chat outputs.