DeepgramMicrosoftAudio matchupSpeech output
Aura-2vs.MAI-Voice-2.1
Which model is better?
MAI-Voice-2.1 is slightly ahead in independent tests, but the lead is not certain.
Pick Aura-2 for
Pick MAI-Voice-2.1 for
Shared tests
Independent evaluators and leaderboards only, the newest shared snapshot per evaluator. With 95 % ranges a lead counts when they do not overlap; for hit rates, when it holds in a 95 % test. Without any uncertainty figure at most “slightly ahead”.
Cekura
MAI-Voice-2.1 slightly aheadMAI-Voice-2.1 is clearly ahead in 0 of 4 subtests and slightly in 2, Aura-2 in 0 clearly and slightly in 1. 1 is level.
Cekura TTS hard-part accuracy: codes and addresses
as of October 19, 2026
Cekura TTS hard-part accuracy: numbers and shorthand
as of October 19, 2026
Cekura TTS hard-part accuracy: medical terms
as of October 19, 2026
Show all 4 subtests
| Subtest | Aura-2 | MAI-Voice-2.1 | Verdict |
|---|---|---|---|
| Cekura TTS hard-part accuracy: codes and addresses | 87.0 % | 67.0 % | Aura-2 slightly ahead |
| Cekura TTS hard-part accuracy: numbers and shorthand | 71.0 % | 85.0 % | MAI-Voice-2.1 slightly ahead |
| Cekura TTS hard-part accuracy: medical terms | 86.0 % | 99.0 % | MAI-Voice-2.1 slightly ahead |
| Cekura TTS hard-part accuracy: read from context | 75.0 % | 73.0 % | level |
Source · latest October 19, 2026
All data side by side
Prices are verbatim from the provider, because audio models bill per minute, character or token.
| Aura-2 | MAI-Voice-2.1 | |
|---|---|---|
| Task | Speech output | Speech output |
| Developer | Deepgram | Microsoft |
| Status | Active | Preview |
| Released | n/a | October 1, 2026 |
| Weights | proprietary | proprietary |
| Runs locally | no | no |
| Price | n/a | $22 per 1M characters |
| Popularity (out of 100) | 44 | 1 |