xcLee001/SonicVale

一个开源的多角色、多情绪 AI 配音生成平台，支持小说、剧本、视频等内容的自动配音与导出。

What it solves

SonicVale (音谷) is designed to automate the process of creating multi-character, multi-emotion voiceovers for novels, scripts, and videos. It eliminates the manual effort of assigning voices and emotions to individual lines of dialogue in long-form content.

How it works

The platform integrates Large Language Models (LLMs) and Text-to-Speech (TTS) engines to transform text into audio. It uses LLMs to automatically split dialogue and assign roles, and leverages the Index-TTS-2.0 engine for emotional voice synthesis. Users can import text, manage a library of character voices, bind specific emotions to roles, and perform precise audio editing (such as adding silence or deleting segments) before exporting the final result.

Who it’s for

Content creators, authors, and video producers who need high-quality, emotionally expressive AI voiceovers for narrative-driven content with multiple characters.

Highlights

  • Automated Dialogue Splitting: Uses customizable prompts and LLMs to break down scripts into character-specific lines.
  • Multi-Emotion Support: Integrates Index-TTS-2.0 to provide nuanced emotional tones for different characters.
  • Character Management: Features a dedicated role library to maintain consistent voice bindings across multiple chapters.
  • Precision Audio Editing: Allows users to manually delete audio fragments or insert silence for better pacing.
  • Flexible Integration: Supports any LLM compatible with the OpenAI API protocol.

Related

  • Project
    jianchang512/pyvideotrans

    pyVideoTrans is an open‑source Python tool that automatically translates videos by chaining speech‑to‑text, AI translation, and text‑to‑speech (with optional voice cloning). It offers a GUI, WebUI, CLI, and Docker deployment, works with local models (e.g., Faster‑Whisper, F5‑TTS) or cloud APIs, and includes utilities for subtitle editing, speaker diarisation, and GPU acceleration.

  • Project
    QwenAudio/CosyVoice

    CosyVoice is an advanced, LLM-based text-to-speech system designed for high-quality, zero-shot multilingual speech synthesis and voice cloning.

  • Project
    jamiepine/voicebox

    An open-source, local-first AI voice studio for voice cloning, text-to-speech, and system-wide dictation, featuring integration with MCP-aware AI agents.

  • Project
    dramaclaw/dramaclaw

    DramaClaw is a self‑hosted, source‑available AI‑driven production platform that turns scripts into short films. It offers a node‑based “infinite canvas” (XiaHua) for free‑form generation of images, video, audio and 3‑D sets, plus a structured “Series” pipeline (XiaJi) that ingests a manuscript, plans episodes, writes scripts, creates story‑boards, synthesises voice‑overs, composes video and exports the final cut. An integrated AI assistant (Xia Director) can drive both tracks. The system runs via Docker, routes all heavy model inference through an OpenAI‑compatible gateway (official or custom), and is licensed under Elastic 2.0, allowing commercial use with a required credit badge.

  • Project
    Xerophayze/TTS-Story

    TTS‑Story is a web UI for multi‑voice audiobook creation. It lets you tag speakers, design or clone voices (via Google Gemini Cloud or many local models), run a queue of TTS jobs, review audio in‑app, and export MP3/M4B. Supports 20+ engines, GPU/CPU isolation, bulk voice casting, directed‑narration, and optional LLM text‑prep.