IndexTTS: An Industrial-Level Controllable and Efficient Zero-Shot Text-To-Speech System

Papers with Code Trending Papers
zero-shot-text-to-speech voice-cloning hybrid-character-pinyin vector-quantization open-source naturalness inference-speed

Summary

IndexTTS is an enhanced text-to-speech system that combines XTTS and Tortoise models with hybrid character-pinyin modeling and optimized vector quantization, achieving improved naturalness, controllable pronunciation, and faster inference than existing open-source TTS systems.

Recently, large language model (LLM) based text-to-speech (TTS) systems have gradually become the mainstream in the industry due to their high naturalness and powerful zero-shot voice cloning capabilities.Here, we introduce the IndexTTS system, which is mainly based on the XTTS and Tortoise model. We add some novel improvements. Specifically, in Chinese scenarios, we adopt a hybrid modeling method that combines characters and pinyin, making the pronunciations of polyphonic characters and long-tail characters controllable. We also performed a comparative analysis of the Vector Quantization (VQ) with Finite-Scalar Quantization (FSQ) for codebook utilization of acoustic speech tokens. To further enhance the effect and stability of voice cloning, we introduce a conformer-based speech conditional encoder and replace the speechcode decoder with BigVGAN2. Compared with XTTS, it has achieved significant improvements in naturalness, content consistency, and zero-shot voice cloning. As for the popular TTS systems in the open-source, such as Fish-Speech, CosyVoice2, FireRedTTS and F5-TTS, IndexTTS has a relatively simple training process, more controllable usage, and faster inference speed. Moreover, its performance surpasses that of these systems. Our demos are available at https://index-tts.github.io.
Original Article
View Cached Full Text

Cached at: 06/21/26, 04:32 AM

Paper page - IndexTTS: An Industrial-Level Controllable and Efficient Zero-Shot Text-To-Speech System

Source: https://huggingface.co/papers/2502.05512

Abstract

IndexTTS, an enhanced text-to-speech system combining XTTS and Tortoise models, offers improved naturalness, enhanced voice cloning, and controllable usage through hybrid character-pinyin modeling and optimized vector quantization.

Recently, large language model (LLM) based text-to-speech (TTS) systems have gradually become the mainstream in the industry due to their highnaturalnessand powerfulzero-shot voice cloningcapabilities.Here, we introduce theIndexTTSsystem, which is mainly based on theXTTSandTortoisemodel. We add some novel improvements. Specifically, in Chinese scenarios, we adopt a hybrid modeling method that combinescharactersandpinyin, making the pronunciations of polyphoniccharactersand long-tailcharacterscontrollable. We also performed a comparative analysis of theVector Quantization (VQ)withFinite-Scalar Quantization (FSQ)for codebook utilization of acoustic speech tokens. To further enhance the effect and stability of voice cloning, we introduce aconformer-based speech conditional encoderand replace the speechcode decoder withBigVGAN2. Compared withXTTS, it has achieved significant improvements innaturalness,content consistency, and zero-shot voice cloning. As for the popular TTS systems in the open-source, such asFish-Speech,CosyVoice2,FireRedTTSandF5-TTS,IndexTTShas a relatively simple training process, more controllable usage, and faster inference speed. Moreover, its performance surpasses that of these systems. Our demos are available at https://index-tts.github.io.

View arXiv pageView PDFProject pageGitHub21.3kautoAdd to collection

Get this paper in your agent:

hf papers read 2502\.05512

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper24

#### IndexTeam/IndexTTS-2 Text-to-Speech• UpdatedJan 20 • 14.8k • 735 #### IndexTeam/Index-TTS Text-to-Speech• UpdatedApr 27, 2025 • 181 • 150 #### taraskurtizan/IndexTTS-2 Text-to-Speech• Updatedabout 12 hours ago • 2 #### Toxzic/indextts-colab UpdatedSep 19, 2025 Browse 24 models citing this paper## Datasets citing this paper1

#### echodict/index-tts

Spaces citing this paper40

Browse 40 spaces citing this paper## Collections including this paper4

Similar Articles

kyutai-labs/pocket-tts

GitHub Trending (daily)

Kyutai releases Pocket TTS, a lightweight text-to-speech model that runs efficiently on CPUs with 100M parameters, low latency, and voice cloning, supporting multiple languages.