Four models. One API.
Speech-to-text for recordings and live audio, speaker diarization and speaker ID, built to stay accurate in real conversations.
Async speech-to-text
a5s-async-v1
A5S Async
Transcripts with speaker labels for meetings, calls, interviews and media.
#1 of 15on VoiceArena's Diarization-Bench
Speaker identificationaircaps-speaker-id-v1
Speaker ID
Recognize enrolled voices, even in noisy environments.
3.11% EERacross clean speech, noise and music
Speaker diarizationaircaps-diarization-v1
Diarization
Who spoke when, across meetings, calls and broadcasts.
12.03% DERlowest of 11 systems across 7 datasets
Streaming speech-to-textDeployments only
A5S Small V2
Streaming speech-to-text for noisy, far-field audio, deployed in your VPC, on premises, air-gapped or on device.
26.27% WERon noisy, far-field audio, ahead of 8 cloud systems
Streaming speech-to-text
a5s-streaming-v1
A5S Streaming
Frontier accuracy streaming speech-to-text with speaker labels, in English and 8 other languages.
Coming soonEarly access for select organizations
Request access
Which one?
A5S Async and Speaker ID are on every account. Diarization is enabled on request. A5S Small V2 Streaming is offered for VPC, on-premise and on-device deployments.
| You have | Use |
|---|---|
| Recordings, uploads, audio and video files | A5S Async |
| The most accurate speaker-labeled transcript of a recording | A5S Async |
| Which known person is speaking: names instead of speaker numbers | Speaker ID |
| Only who spoke when, or speakers for a transcript you already have | Diarization (on request) |
| Streaming speech-to-text in your own cloud, on premises or on a device | A5S Small V2 (talk to sales) |
| Live captions with speaker labels | A5S Streaming (coming soon; early access on request) |
Hear what other models miss.
Run AirCaps AI on your own audio. Every account includes 5 free hours.