Lowest-latency real-time / streaming transcription API
For live, low-latency streaming, ElevenLabs Scribe v2 and AssemblyAI Universal-3 Pro both publish around 150 ms — the lowest here — though ElevenLabs runs streaming on a separate realtime mode. Deepgram Nova-3 markets sub-300 ms and is built streaming-first with per-second billing. These numbers mix measured P50 with vendor claims and aren't strictly comparable, so weigh them loosely. A light estimate from public sources, not a first-hand latency benchmark.
DefaultElevenLabs Scribe v2laagste latencyElevenLabs Scribe v2
Provider offerings
| Offering | Price ($/1000 min) | WER | Langs | Latency | Capabilities |
|---|---|---|---|---|---|
| ElevenLabs Scribe v2ElevenLabs | 3.67 | 2.2% | 90 | 150 ms | diarizationtimestampsvocab |
| AssemblyAI Universal-3 ProAssemblyAI | 3.5 | 3.1% | 99 | 150 ms | diarizationtimestampsvocab |
| Deepgram Nova-3Deepgram | 4.3 | 5.2% | 50 | 300 ms | diarizationtimestampsvocab |
| Speechmatics EnhancedSpeechmatics | 6.7 | 4% | 70 | 500 ms | diarizationtimestampsvocab |
| OpenAI gpt-4o-transcribeOpenAI | 6 | 4% | — | — | — |
| Google Gemini 3 FlashGoogle | 1.92* | 2.9% | — | — | — |
* token-/credit-priced — the headline understates real per-unit cost, so it is excluded from the cheapest ranking.
Sources
- Artificial Analysis — Speech to Text2026-06-19
- Open ASR Leaderboard2026-06-19