AirCaps AI

An AssemblyAI alternative for voice agents.

AirCaps Live 1 returns the final transcript a median 38 ms after the caller stops speaking, against 99 ms for AssemblyAI Universal-3.6 Pro, and makes 15% fewer word errors (3.65% against 4.31% WER on 9,205 voice-agent clips). Its built-in turn detection ends 6× fewer turns while the caller is still mid-thought (3.6% against 21.3%). It costs $0.20 per audio hour; AssemblyAI Universal-3.6 Pro lists at $0.45.

38 msmedian time to final transcript; Universal-3.6 Pro takes 99 ms, and 4.1× as long at p95
3.65%word error rate on 9,205 voice-agent clips; Universal-3.6 Pro 4.31%
6×fewer callers cut off mid-thought than Universal-3.6 Pro
$0.20per audio hour; Universal-3.6 Pro lists at $0.45

Side by side.

The same audio, streamed in real time to every model. The better result in each row is highlighted.

What changes when you switch.

Fewer word errors

3.65% WER on 9,205 voice-agent clips, against 4.31% and 4.42% for Universal-3.6 Pro and Universal-3.5 Pro. Live 1 is also lower on each of the four public sets in the table.

A sooner final transcript

A median 38 ms from the end of speech to the final transcript, 52 ms at p95. Universal-3.6 Pro takes 99 ms, and 211 ms at p95.

Fewer callers cut off

On Pipecat's Smart Turn v3.2 set, Live 1 ended 3.6% of unfinished turns too early; Universal-3.6 Pro ended 21.3%. Live 1 also calls the end of a turn sooner: 111 ms against 390 ms.

A lower price

$0.20 per audio hour, with 5 free hours on every account. Universal-3.6 Pro lists at $0.45.

Questions.

Is AirCaps Live 1 a good alternative to AssemblyAI Universal-3.6 Pro?

AirCaps Live 1 returns the final transcript a median 38 ms after the caller stops speaking, against 99 ms for AssemblyAI Universal-3.6 Pro, and makes 15% fewer word errors (3.65% against 4.31% WER on 9,205 voice-agent clips). Its built-in turn detection ends 6× fewer turns while the caller is still mid-thought (3.6% against 21.3%). It costs $0.20 per audio hour; AssemblyAI Universal-3.6 Pro lists at $0.45.

How do I switch from AssemblyAI to AirCaps Live 1?

Stream 16 kHz mono PCM to the Live 1 WebSocket and read the transcript messages. Turn on end_of_turn to let the model end turns, or send finalize from your own VAD; both work in the same session. The AirCaps Live 1 guide at aircaps.ai/docs/guides/live-1 has the whole protocol.

How does its turn detection compare with AssemblyAI Universal-3.6 Pro?

On Pipecat's Smart Turn v3.2 English set (5,367 clips, 2,718 of them unfinished turns), Live 1 ended 3.6% of unfinished turns too early and 93.5% of its turn endings were right; AssemblyAI Universal-3.6 Pro ended 21.3% too early, with 74.8% right. On LiveKit EOT Bench (7,655 turns), Live 1 called the end of a turn a median 111 ms after the caller stopped; AssemblyAI Universal-3.6 Pro took 390 ms.

How was this measured?

9,205 English clips (154,661 words) from Pipecat's STT benchmark, Coval's public set, LiveKit EOT Bench, WildASR and PriMock57, run on November 19, 2026. Every model streamed the same audio in real time from one machine in AWS us-east-2. Time to final transcript runs from the end of speech, when the last audio and the finalize message have been sent, to the final transcript, network included.

How much does AirCaps Live 1 cost?

$0.20 per hour of audio you stream, billed per millisecond, with 5 free hours on every account. AssemblyAI Universal-3.6 Pro lists at $0.45; AssemblyAI Universal-3.5 Pro lists at $0.45.

Does it work with Pipecat and LiveKit?

Yes. It runs over a WebSocket, so it fits a Pipecat pipeline or a LiveKit agent as the speech-to-text step.

Which languages does it support?

English, with multilingual support coming soon.

Try it on your calls.

AirCaps Live 1 is on every account, with 5 free hours. Read the AirCaps Live 1 guide.