XTTS v2

Text to speech · by Coqui
🔓 Open-weight

Ranking

Text to speech
#81 of 91
914 ELO
324 ELO behind the leader (Qwen-Audio-3.0-TTS-Plus)

Run it yourself

XTTS v2 has open weights you can download and run on your own hardware. ~4 GB VRAM recommended · License: Coqui Public Model. Works with Coqui TTS.

Find the weights on Hugging Face ↗ Build a local stack 🧱

Other Text to speech models

Qwen-Audio-3.0-TTS-Plus#1 · AlibabaSimba 3.2#2 · SpeechifyAIGemini 3.1 Flash TTS#3 · GoogleSonic 3.5#4 · CartesiaRealtime TTS 1.5 Max#5 · InworldLightning V3.1 Pro (Jul 2026)#6 · Smallest.aiRealtime TTS-2 - Research Preview#7 · InworldLuna TTS#8 · VUI Labs
How this is ranked: ELO scores from blind human preference (people compare two outputs and pick the one they prefer) — rankings by Artificial Analysis. Open-weight means the model can be downloaded and run on your own hardware; where we cannot confirm it, we mark the model unverified rather than guess. See the full media model leaderboard.