146 ms median and 213 ms P95 time to first audio on Coval (30 days to October 22, 2026): Fast enough speed tier.
Benchmark summary: Fast enough speed tier on Coval's P95 (fast enough for most agents), a tie within 10 ms with Palabra TTS v1 on slow turns. On the Coval and Artificial Analysis frontier: no other model is both faster and rated higher.
Fish Audio Drama 3 Preview is an open-weights text-to-speech (TTS) API from Fish Audio, measured by two independent benchmarks: Artificial Analysis and Coval.
Speed tier
Fast enoughP95 from 138 ms to 263 ms. Fast enough for most agents.
Data read October 22, 2026. No combined score: methodology
Gradium TTS vs Fish Audio Drama 3 Preview Fastest TTS APIs, by speed tierOn Coval's 30-day window ending October 22, 2026 (922 runs from US East), Fish Audio Drama 3 Preview starts speaking after 146.1 ms on a typical turn (median) and 213.2 ms on a slow turn (P95, the slowest one in twenty). That P95 is 10th lowest of the 32 models Coval measures, which puts it in the Fast enough tier: fast enough for most agents. On the median it is a tie (gap under 10 ms) with Inworld Realtime TTS-2.
In the Artificial Analysis Speech Arena it holds 1,248 ±15 Elo (95% interval), 9th highest of the 98 models Artificial Analysis rates; scores whose intervals overlap are a tie. In its Customer Service category, Artificial Analysis places it 8th, at 1,301 Elo.
Speko has not scored its pronunciation of numbers and dates.
Artificial Analysis lists $15. List prices, as each source prints them.
Palabra TTS v1 is a proprietary text-to-speech (TTS) API from Palabra, measured by two independent benchmarks: Coval and Speko.
Inworld Realtime TTS-2 is a proprietary text-to-speech (TTS) API from Inworld, measured by three independent benchmarks: Artificial Analysis, Coval and Speko.
ElevenLabs Flash v2.5 is a proprietary text-to-speech (TTS) API from ElevenLabs, measured by three independent benchmarks: Artificial Analysis, Coval and Speko.