Favicon of YoVoice

YoVoice

A local AI audio studio for macOS and Windows that generates speech, clones voices and creates music, with offline use after model downloads.

Screenshot of YoVoice website

YoVoice is a desktop audio and music studio for creators making narrated stories, voiceovers or songs on their own computer. It combines generation with editing and arrangement tools.

  • Speech and character voices: Turn text or subtitles into narration, assign voices to characters, clone reference recordings or design a voice from a description. Emotion and delivery controls depend on the model.
  • Songs and instrumentals: Generate music from style descriptions and lyrics with ACE-Step, or create vocal songs with YuE2. Compare saved takes and trim audio before export.
  • Multi-track arrangements: Write separate musical parts or import MIDI, render them with General MIDI sound libraries, and export MIDI and individual tracks. Score rendering doesn't require a GPU.
  • Agent integration: Give AI agents access through a CLI, MCP or an authenticated HTTP API. An optional assistant can write lyrics and arrangements through Ollama or providers such as OpenAI and Anthropic.

Speech backends include IndexTTS, VoxCPM2, Qwen3-TTS and Kokoro. Generation works offline once models and sound libraries are downloaded. Text, reference recordings and finished work stay on your machine; optional AI requests go to the service you choose.

The software is open source under Apache 2.0. It runs on macOS 14 or later with Apple Silicon and Metal, or 64-bit Windows 10/11 with CPU or NVIDIA CUDA support. Full-song generation recommends 12 GB or more of VRAM, or 16 GB or more on Apple Silicon.

Voice and music models have separate license terms.

Similar to YoVoice