Fish Audio S2 Pro
Audio Available Full comparison ↗ fish-audio/s2-pro · by Fish Audio
· neural-tts
Pricing — 1 offering(s)
Text characters
- $15.00 / 1M characters
Current
2026-07-30 → present
Fish Audio's native billing unit is UTF-8 BYTES, not characters — the pricing page states "$15.00 / M UTF-8…
Fish Audio's native billing unit is UTF-8 BYTES, not characters — the pricing page states "$15.00 / M UTF-8 bytes", identical to S1 and S2.1-Pro. For English/Latin-script text 1 character = 1 byte; for CJK and other multi-byte scripts actual cost per character can run up to 3-4x this rate. Recorded as per_1m_characters (closest schema unit) since the registry has no byte-based unit; effective_from is the date this entry was sourced, not a confirmed price-change date.
Showing the active price and any recorded history. Full pricing history is available via the paid API — see API docs.
See how Fish Audio S2 Pro fits into a cost-aware routing setup
See how →Capability profile
How Fish Audio S2 Pro rates across core capability dimensions, with the task-level evidence behind each rating.
Natural-language [bracket]-tag emotion control (not limited to a fixed tag set). Fish Audio positions this as a production-grade model, one step below S2.1 Pro's 'recommended' status.
80+ languages — broader than ElevenLabs' models (29-70) and well above Fish's own S1 (13).
Multi-speaker dialogue support plus voice cloning from reference audio; access to Fish Audio's platform-wide voice library.
100ms time-to-first-audio guarantee, per Fish Audio's own docs — a contractual SLA, not an independently measured figure.
Voice cloning from reference audio, same platform-wide capability as S1/S2.1 Pro.
Benchmarks
| Benchmark | Score | Config | Source |
|---|---|---|---|
| Time-to-first-audio guarantee (vendor SLA) | 100 ms | Fish Audio's own contractual production-tier guarantee, not an independently measured latency benchmark. | source ↗ |
Operator guidance
Fish Audio's own docs label S2 Pro "Previous S2 Model" and recommend S2.1 Pro for new production integrations — same $15/1M-byte price, with improved quality, latency, throughput, and broader language coverage (83 vs 80+ languages). S2 Pro remains fully supported and priced, not deprecated, so existing integrations don't need to migrate urgently.
Use cases
- Production TTS with multi-speaker dialogue and open-ended emotion tagging
- Latency-sensitive applications relying on the 100ms TTFA guarantee
- Existing integrations on S2 Pro not yet migrated to S2.1 Pro
Limitations
- Fish Audio recommends S2.1 Pro over S2 Pro for new production integrations
- 100ms TTFA figure is a vendor SLA commitment, not an independently measured benchmark
- Fish Audio's native billing unit is UTF-8 bytes, not characters — see registry pricing notes