Fish Audio
Fish Audio Drama
Current
3.0
shipped 3w · 16 Oct 2026
35B MoE emotive speech model with 16-character multi-speaker synthesis.
Specification
- Architecture
- 35B MoE (3B active)
- Emotions
- Bracket tags ([whispers], [laughs])
- Benchmark
- #1 Artificial Analysis (93.2%)
- Weights
- Open (Fish Audio Community v1.0)
- Origin
- China
Modalities: voice, text
Fish Audio Drama — 6-Axis Capability Radar
HEURISTIC ESTIMATEThese scores are rough estimates from category presets, catalog specs, and keywords in release notes. They are not measured benchmarks or test results. Catalog prices and context windows may refer to a different model variant.
Hover or tap an axis to inspect its estimate
Shipping Cadence & Next Drop Forecast
This is an uncalibrated estimate from limited catalog history. A ratio of 100% means one estimated interval has elapsed; it is not a probability. Next-release dates and versions remain unconfirmed.
11 days
±7d variance26 days
Shipped on 2026-10-1633.2 drops/yr
1 observed intervals · limited history2026-10-27
Next version: UnannouncedOptional AI cadence commentary
Summarize the recorded history and its limitations. Future releases remain unconfirmed.
Model Version Capability Delta
AI-synthesized architectural jumps, parameter scaling, and capability shifts across releases.
Ready to analyze Fish Audio Drama v2.0 vs v3.0
Click Synthesize Diff to analyze latent space adjustments, resolution leaps, parameter shifts, and migration caveats via Gemini 3.8 Flash.
Version history
2 releases tracked
3.0
major16 Oct 2026 · 3wFish Audio drops Drama 3: 35B MoE emotive speech model with 16-character multi-speaker synthesis.
- Directable performance via natural language bracket tags ([whispers], [laughs], [anger], [crying]).
- Multi-character dialogue handling up to 16 distinct speaker tags in a single request.
- Ranked #1 on Artificial Analysis Pronunciation Robustness benchmark (93.2% accuracy).
Recorded announcementSpecific source link · unverified ↗2.0
minor5 Oct 2026 · 1moFish Speech 2.0 introduces sub-100ms streaming TTS.
- Zero-shot voice cloning from 3 seconds of audio.
Date basis unconfirmedSource needed