IndexTTS-2.5

All projects

Apps & Tools

A zero-shot multilingual text-to-speech model with emotion and speed control.

byAI Search

Playable

IndexTTS-2.5 media is blocked

Allow external media to connect to the provider and play this content.

Description

IndexTTS-2.5 is a zero-shot voice cloning model supporting Chinese, English, Japanese, Spanish, and Arabic. It features an autoregressive GPT backbone with emotion control disentangled from timbre and controllable speaking speed, requiring roughly 6 GB of VRAM for inference.

Descriptions, tags, and model credits may be AI-generated or inferred from public sources and can be incomplete or wrong. Learn more.

Details

Why this is interesting: Impressive zero-shot voice cloning capabilities combined with precise control over emotion and speed across five…
Attributed to: @aisearch
Built with: Seedance 2.5 Seedance 2.0 w2v-BERT 2.0
Links: huggingface.co
View on YouTube
Tags: #tts #voice-cloning #multilingual #audio-generation

Discovered: YouTube
First posted: YouTube · Aug 16, 2026
Listed: Added 5d ago
Build: Strong
Models: Moderate

Prompt

Animal Liberation and the Royal Society for the Prevention of Cruelty to Animals (RSPCA) are again calling for the mandatory installation of CCTV cameras in all Australian abattoirs.

Source

IndexTTS-2.5

Description

Timeline

IndexTTS-2.5

Description

Timeline