Best transcription API for podcasts
Default pickElevenLabs Scribe v2
For podcast episodes (long multi-speaker recordings) ElevenLabs Scribe v2 is the strongest default: the highest measured accuracy on the Artificial Analysis leaderboard, with full speaker diarization and word-level timestamps, at $3.67 per 1000 minutes. AssemblyAI Universal-3 Pro is the cheapest directly-comparable option at $3.50 per 1000 minutes, with the same diarization and timestamp support for long batch audio. A light estimate from public benchmarks and pricing, August 2026.
DefaultElevenLabs Scribe v2met diarisatieElevenLabs Scribe v2goedkoopsteAssemblyAI Universal-3 Pro
Provider offerings
| Offering | Price ($/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 | —* | 4% | 70 | 500 ms | diarizationtimestampsvocab |
| OpenAI gpt-transcribeOpenAI | 4.5 | 3.3% | — | — | vocab |
| 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 Text2026-08-20
- Open ASR Leaderboard2026-06-19