New: Medical-1 Flash · $0.18 / audio-hour →

Voice AI built for healthcare.

Add medical speech-to-text to a prototype, a startup product or a healthcare AI platform. Version 1.1 made 21% fewer medical-term errors than v1 on our live-transcription benchmark.

For clinicians who code · Healthtech startups · Healthcare AI companies

25 free audio-hours monthlySelf-serve BAANever used for model training
Consultation audio
“Start amoxicillin 500 mg and repeat HbA1c at review.”
✓ Drug and dosage preserved
#1of 20 batch APIs · medical-term error
#1of 13 realtime models · medical-term error
21% fewerrealtime medical-term errors · v1.1 vs v1

Same 1,513 clinical recordings for every model. What's new in v1.1 →

Built for the workflow

Consultations, dictation and patient calls.

Use the same medical speech layer across live conversations, recorded notes and local workflows.

Consultations

Ambient documentation

Capture speakers, timing and the medical language needed for notes and downstream agents.

Dictation

Clinical text entry

Give clinicians a faster input method inside EHR, dental, therapy and specialty workflows.

Patient calls

Voice workflows

Feed confirmed medical text into triage, follow-up and care-navigation logic.

Local AI

Private on-device speech

Run the open model in your own application when audio cannot leave the environment.

Choose how it runs

Hosted API or open model.

Hosted API

Start with 25 free hours.

Use batch and live transcription with medical vocabulary, speakers and timestamps.

  • No card required
  • Self-serve BAA and DPA
  • Pay only when usage grows
Open model

Run it in your environment.

Deploy omi-medical-edge-1 on Apple Silicon, NVIDIA CUDA or CPU—or try its English hosted endpoint first.

  • CC-BY-4.0 weights
  • MIT runtime
  • Published results for each platform
Benchmark results

Every layer of the medical record, measured.

Compare medical terms, drug names and doses. Then see how quickly the final transcript arrives.

Medical-term error ↓

Omi Medical v1.10.87%
ElevenLabs Scribe v2 Medical1.08%
Microsoft MAI-Transcribe-21.18%
Deepgram Nova-3 Pharma1.22%

Dose / unit matches ↑

Omi Medical v1.196.81%
ElevenLabs Scribe v2 Medical86.17%
Microsoft MAI-Transcribe-285.11%
Deepgram Nova-3 Pharma80.85%

Batch response · median ↓

Omi Medical v1.1131 ms
Microsoft MAI-Transcribe-2200 ms
ElevenLabs Scribe v2 Medical279 ms
Deepgram Nova-3 Pharma373 ms

Selected batch results. Every API timed from the US (Ohio), October 2026. Speed: clips under 30 seconds. Full comparison →

Realtime: 1.04% medical-term error, lowest of 12 tested. Final transcript 86 ms median after client-signalled speech end (169 ms p95), measured in the US. See realtime results →

Research and evidence

See the work behind the claims.

Open benchmarks, published model results and reproducible evaluation.

Flagship model

Omi Medical v1.1

What changed in v1.1: realtime accuracy, speed and safeguards for difficult audio.

Read the model report →
New model

Omi Medical-1 Flash

Fast English batch transcription in the US at $0.18 per audio-hour.

Read about Flash →

Start with your own audio.

Use the hosted API or download the open model.