Project AURIONAI Use-Cases Research & Innovation
Fish Audio
speech synthesis

Fish Audio S2 Pro

Fish Audio open TTS model; commercial use needs a licence.

Fish AudioFish AudioUSA
Overview of Fish Audio S2 Pro: fine-grained control, multi-speaker multi-turn generation, fast streaming and long audio.
Levelmodel
Classspeech synthesis
Parameters5B
Contextnot in the notes
ReleasedMar 2026
Opennessopen weights restricted
Runs onsingle datacentre GPU
Statusactive

Lineage

successor
Fish Speech to Fish Audio S2 Pro: successor, low confidenceFish SpeechFish SpeechFish AudioFish Audio S2 ProFish Audio S2 ProFish Audio, Mar 2026
Fish AudioFish Speech is the parent (successor)low confidencenotes
Where this comes from
Parent and child relations from model cards, Hugging Face base_model fields and the genealogy tables of the research notes. The tree shows two generations each way, 2 models. Confidence and evidence are the notes' own.

R&D Topics

Large-scale zero-shot text-to-speechproduced by

Identity

Developer
Fish Audio
Family
not in the notes
Released
Mar 2026 sourcesregistry date
Weights published
–
Record checked
10 Oct 2026
Where this comes from
Developer and dates from the research notes. Release date is the vendor's public announcement; weights date is when open weights were published.

Architecture and Size

Architecture
autoregressive Dual-AR decoder-only plus RVQ codec
Total parameters
5B sources
Active parameters
–
Checkpoint parameters
–
Note
About 5B
Context length
not in the notes
Input
text
Output
audio
Where this comes from
Parameter counts as the developer states them. Where the checkpoint holds a different total, both are shown (register A16).

Openness and Access

Openness
open weights restricted
Licence
Fish Audio Research Licence sources
Access routes
open weights download
Where this comes from
The licence file governs over registry tags and aggregator labels.

Deployment

Runs on
single datacentre GPU
Footprint
not in the notes
Runtimes
not in the notes
Where this comes from
Deployment tiers and runtimes as named in the notes. Footprint is the notes' figure, not a measurement by AURION.

Indonesia

Indonesian support
listed by developer
Regional languages
not in the notes
Jakarta hosting
unknown
Languages
80+ languages including Indonesian (id)
Where this comes from
Indonesia fields are carried on every model; unknown is an allowed value.

Benchmark Results

The notes record no benchmark result for this model.

Industry Use Cases

No industry use case links this model.

Sources

  1. fishaudio/s2-pro , Hugging Face huggingface.co https://huggingface.co/fishaudio/s2-pro
  2. Fish Audio Raises $52 Million Seed Round To Expand Expressive Voice AI Platform pulse2.com https://pulse2.com/fish-audio-raises-52-million-seed-round-to-expand-expressive-voice-ai-platform/

To cite this page: Project AURION, AI Use-Cases Research & Innovation, "Fish Audio S2 Pro", https://aurion.semai.id/models/fish-audio-s2-pro/, records of October 2026.