AGIdex
agidex/Technologies/Whisper Large v3
AI MODEL

Whisper Large v3

OPEN WEIGHTS
RELEASED 2023-11

Open-weight speech recognition. 2.7% WER on clean audio. Supports 99 languages.

Benchmark coverage · per sub-capability

Each score is anchored to human baseline (100). Source URLs link to the original benchmark, leaderboard, or release note.

03HearingCOGNITIVE
Sub-capabilityQualitySourceScore vs. baselineScore
Speaker Identificationindependentpaperswithcode.com/sota/speaker-verification-on-voxceleb1
94%
Speech Recognitionindependentopenai.com/research/whisper
103%