leemysw/yovoice

Open-source local AI audio & music studio for macOS & Windows. TTS, voice cloning, songs with lyrics, and multi-track scores — 100% offline, zero API fees.

What it solves

yovoice provides a local, open-source alternative to cloud-based text-to-speech (TTS) services. It eliminates per-character costs and privacy concerns by allowing users to generate natural, expressive speech on their own hardware, specifically for narration, voiceovers, and audio content creation.

How it works

The tool integrates several local TTS models (such as IndexTTS, VoxCPM2, OmniVoice, Qwen3-TTS, and Kokoro) using the audio.cpp inference engine. It supports hardware acceleration via Metal (Apple Silicon), NVIDIA CUDA, and Vulkan (Windows). Users can generate speech through a desktop application or a CLI, with features including voice cloning from reference audio, text-guided voice design, and emotion control.

Who it’s for

Content creators, narrators, and developers who need high-quality, controllable voiceovers without relying on cloud APIs or incurring subscription fees.

Highlights

  • Local Execution: No cloud API calls or per-character charges.
  • Diverse Model Support: Compatible with a wide range of models for different needs, from lightweight (Kokoro) to high-parameter models (Qwen3-TTS).
  • Advanced Voice Control: Supports voice cloning, emotion adjustment, and text-guided voice design (creating voices from natural language descriptions).
  • Cross-Platform: Native support for macOS (Apple Silicon) and Windows.
  • Agent Integration: Includes a "Skill" that allows AI agents to automate speech generation via the CLI.

Related

  • Project
    mnemox-ai/tradememory-protocol

    TradeMemory Protocol is an open‑source Python library that records an AI trader’s own past trades, scores them with a five‑factor Outcome‑Weighted Memory model, and can block or reorder future orders before they reach a broker. It syncs fills from venues like Hyperliquid, Alpaca, MT5, and Binance, provides a local MCP server (and an optional proxy brake) for pre‑flight risk checks, and maintains a tamper‑evident SHA‑256‑chained audit log.

  • Project
    milanglacier/minuet-ai.nvim

    Minuet is a Neovim plugin that adds AI‑powered code completion. It talks to any OpenAI‑compatible or FIM‑compatible LLM (OpenAI, Claude, Ollama, Llama‑cpp, etc.) via curl, and can feed suggestions to virtual‑text, nvim‑cmp, blink‑cmp, or Neovim’s built‑in LSP/inline‑completion. It also offers an experimental “next‑edit prediction” (Duet) mode. Install via lazy.nvim or LuaRocks, configure the provider, API key and UI preferences, then trigger completions with the supplied keymaps.

  • Project
    obsei/obsei

    obsei is an open‑source, self‑hosted platform for privacy‑preserving customer‑feedback analytics. It ingests data from many sources, redacts PII, enriches records with LLMs or lightweight decision models, stores results in an encrypted DuckDB, and provides routing, theme clustering (multilingual), a query interface, and a read‑only Studio UI. All processing runs inside your own infrastructure, letting you comply with GDPR, CCPA, etc., while still leveraging AI models you control.

  • Project
    paul-buerkner/brms

    brms is an R package that provides a high‑level, lme4‑style interface for fitting Bayesian generalized (non‑linear) multilevel models using Stan. It supports many response families, flexible priors, and a suite of post‑fit diagnostics and visualizations, enabling users to perform sophisticated Bayesian regression without writing Stan code directly.