AssemblyAI Author Benchmark - streaming - Caller entity classes - Technical - Normalized entity error rate: leaderboard
AssemblyAI Author Benchmark - streaming - Caller entity classes - Technical - Normalized entity error rate — AssemblyAI primary author evaluation, captured 2026-11-16; no fusion with another campaign sharing a corpus. Exact system versions and run ranges retained in source_systems_rosters. Prerecorded: public API defaults, auto language except explicitly pinned Cohere/Azure; models shown only when >=90% of files returned relative to AssemblyAI. Streaming: real-time SDK/WebSocket delivery, fixed language, final transcripts rather than partials, shared normalization. All streaming 0.00 means no data. Language WER/CER >=60 means unsupported. Language-specific boards join all-language and partial-language rosters but exclude incomparable means over different supported languages. Native percentage error; no upper cap of100. Metric: Normalized entity error rate (%). Source: https://www.assemblyai.com/benchmarks
Metric: Normalized entity error rate (%). Source: www.assemblyai.com. Saturation forecast: Rough model projection: around 2026. 22 models tracked. Updated 16 Nov 2026.
Top models
| # | Model | Setting | Score | Overall rank |
|---|---|---|---|---|
| 1 | AssemblyAI Universal-3.6 Pro | 20.9 | ||
| 2 | AssemblyAI Universal-3.5 Pro | assemblyai_author_benchmarks_2026 run | 21.7 | |
| 3 | Zoom Scribe (Live) | assemblyai_author_benchmarks_2026 run | 26.4 | |
| 4 | Soniox Realtime | assemblyai_author_benchmarks_2026 run | 26.5 | |
| 5 | ElevenLabs Scribe v2 | assemblyai_author_benchmarks_2026 run | 27.2 | |
| 6 | Google Gemini 3.5 Live | assemblyai_author_benchmarks_2026 run | 27.4 | |
| 7 | Meta Muse Voice Transcribe | assemblyai_author_benchmarks_2026 run | 30.5 | |
| 8 | Reson8 | assemblyai_author_benchmarks_2026 run | 31.3 | |
| 9 | Speechmatics Enhanced | 31.8 | ||
| 10 | AWS Transcribe | assemblyai_author_benchmarks_2026 run | 33.7 | |
| 11 | xAI Grok Streaming | assemblyai_author_benchmarks_2026 run | 34.3 | |
| 12 | Mistral Voxtral Mini | assemblyai_author_benchmarks_2026 run | 34.3 | |
| 13 | Speechmatics Standard | assemblyai_author_benchmarks_2026 run | 36.2 | |
| 14 | Deepgram Nova-3 | assemblyai_author_benchmarks_2026 run | 39.4 | |
| 15 | Gladia Solaria | assemblyai_author_benchmarks_2026 run | 40 |
Interactive version: theaggregate.ai/benchmark?slug=assemblyai-author-benchmark-streaming-caller-entity-classes-technical-normalized-entity-error-rate · How It Works · Data refreshed daily, snapshot 2026-11-16.