indextts2

Draft to Take beta: local-first AI audio production studio powered by IndexTTS2, Docker, Qwen, OmniVoice, SFX, ambience, and music sidecars.

IndexTTS2 精确停顿控制（[pause:N]）ComfyUI 节点包：句中与句号前后停顿精确调整，平均偏差 13ms；支持批量生成、候选试听验收、SRT 字幕与 seed 复现。Precise pause control ([pause:N]) ComfyUI nodes for IndexTTS2 — waveform-domain pause editing in-sentence and around periods, ±13ms, with batch generation, candidate acceptance and SRT subtitles.

Local-first AI orchestration with a host-authoritative Rust runtime, extension-driven capabilities, and first-party support for local LLM, audio, and video workflows. The hub where everything connects and extends.

Structural delivery rules discovered through controlled experiments with IndexTTS-2

A reviewable, resumable Codex Skill that turns verified creator knowledge into articles, 3–5 minute vertical videos, and seven-platform publishing packages.

AI Agent 可复用的本地语音生成技能 —— 通过 gradio_client 调用本机正在运行的 IndexTTS2 WebUI，复用已加载模型生成口播/配音（不重复加载、不占双份显存）。

LingChat 内置 IndexTTS2 的 Windows AMD ROCm 运行时与官方模型安装器

Archived Draft to Take beta 8 distribution. Current releases and documentation are in JaySpiffy/draft-to-take.

Local-first AI audio production studio for stories, podcasts, immutable A/B Takes, sound design, mastering, and release packages.

Local multi-voice audiobook pipeline for the Shadow Slave web novel - LLM diarization + IndexTTS2 emotional voice cloning on a single 12 GB GPU

macOS SwiftUI app for zero-shot voice cloning with IndexTTS2 and MLX on Apple Silicon — no cloud, no API keys, runs entirely locally

IndexTTS2 fine-tuning pipeline — first public training code for IndexTTS2. 8 known pitfalls from IMDA NSC production runs.