Fish Audio Drama 3

Open-weight expressive TTS with voice cloning

🎙️ Voice & audio NEW THIS WEEK provisional score open source / open weights

/ 100

Fish Audio's open-weight expressive TTS model with voice cloning and multi-speaker output, published on Hugging Face (Oct 2026).

You voice game characters, podcasts or trailers and have big GPUs.

You lack a large NVIDIA GPU; use a hosted TTS instead.

Score breakdown

Job fit15/30

Ease / time-to-first-result7/15

Value12/15

Momentum10/15

Openness & portability9/10

Trust & maturity6/10

Builder/gamer bonus4/5

Initial editorial estimate from public info; not yet hands-on tested. Provisional (new release, under 30 days). How we score →

Pricing: Open weights (check license); heavy GPU needed locally
Unverified. Unverified: editorial summary. We couldn't confirm this on an official pricing page (JS-rendered, no public pricing, or usage-based) as of Oct 18, 2026.
Free tier: Yes (unverified)
Platforms: Self-host
Open source: Yes
Official URL: https://huggingface.co/fishaudio/drama-3
URL verified: Oct 18, 2026 (curl HTTP GET)
Last reviewed: Oct 18, 2026 by agent
Sources: https://huggingface.co/fishaudio/drama-3