Vendor page
Meet Scribe
ElevenLabs, 2025
Read the originalRead on October 17, 2026
- What it is
- ElevenLabs' launch page for its Scribe speech-to-text model. It describes the model's features and cites its own benchmark charts.
- How much weight it carries
- A vendor's own launch claims. The benchmark charts are images, and the page does not give a test method, so treat the comparisons as the vendor's claims until independently checked.
Facts taken from this source
- ElevenLabs says its Scribe speech-to-text API returns structured JSON transcripts with speaker diarization and word-level timestamps.
“to get structured JSON transcripts with speaker diarization and word-level timestamps” (Developer paragraph)
- ElevenLabs says Scribe outperforms Gemini 2.0 Flash, Whisper Large V3 and Deepgram Nova-3 on the FLEURS and Common Voice benchmarks across 99 languages.
“In FLEURS & Common Voice benchmark tests across 99 languages, it consistently outperforms leading models like Gemini 2.0 Flash, Whisper Large V3 and Deepgram Nova-3.” (Introduction)
Pages that rest on this source
- Timestamps and Speaker Labels in TranscriptsGuide
- Can Whisper identify different speakers?Question
- How do you transcribe audio with multiple speakers?Question
- Is ElevenLabs speech to text better than Whisper?Question
- Which speech-to-text APIs can identify speakers?Question
- Is Whisper more accurate than Google Speech-to-Text?Question