Can it transcribe healthcare call-center calls in real time?

Yes — Universal-3.6 Pro Realtime with Medical Mode at a 307 ms median time to final transcript, with unlimited concurrency for high volume.

Does it capture both clinical and administrative details?

Yes — medication names and symptoms alongside scheduling and insurance terms, in one transcript.

Can I add sentiment and topic analytics on top?

Yes — Speech Understanding adds sentiment, topic detection, and PII/PHI redaction.

Is it HIPAA-appropriate?

AssemblyAI is a business associate under HIPAA and signs a standard BAA; PHI redaction; SOC 2 Type 2.

How does it scale, and what does it cost?

Pay-as-you-go per second, no concurrency caps; Medical Mode $0.15/hr add-on.

How do I enable Medical Mode for a medical call center?

Add domain: "medical-v1" to the request.

Medical call centers

Patient-access calls, transcribed with clinical accuracy

Front-office calls mix scheduling, insurance, and clinical questions — and the clinical details are exactly what general STT drops. Universal-3.5 Pro with Medical Mode keeps medication names, symptoms, and member details intact across high call volume, live on Universal-3.6 Pro Realtime or post-call.

Try Medical Mode free on your audio Read the docs

Accuracy that holds across thousands of concurrent calls

Missed Entity Rate 3.2%

Missed Entity Rate on clinical conversations with Medical Mode.

vs. base model ~20%

Fewer missed medical entities than the base model without Medical Mode.

vs. base model 87%

Fewer medical entity errors than the base model.

Benchmarked providers #1

Lowest Missed Entity Rate vs. Deepgram, Speechmatics, AWS, and Google.

See the medical benchmarks

Activation

Live transcription for patient-access agents

Stream live calls for agent assist and QA — one WebSocket, one parameter, unlimited concurrency.

from assemblyai.streaming.v3 import StreamingClient, StreamingParameters

client = StreamingClient(api_key="YOUR_API_KEY")

# Medical call center: scheduling + clinical questions
client.connect(StreamingParameters(
    speech_model="universal-3-6-pro",
    domain="medical-v1",
    keyterms_prompt=["referral", "prior authorization", "amoxicillin", "co-pay"],
))

Clinical + administrative terms

Medications, symptoms, insurance and scheduling vocabulary in the same call.

Universal-3.5 Pro

Real-time agent assist

Stream transcripts for live agent assist and QA at a 307 ms median time to final transcript.

Universal-3.6 Pro Realtime

Scale + analytics

Unlimited concurrency plus Speech Understanding for sentiment, topics, and PHI redaction.

Speech Understanding

Common questions

: Yes — Universal-3.6 Pro Realtime with Medical Mode at a 307 ms median time to final transcript, with unlimited concurrency for high volume.
: Yes — medication names and symptoms alongside scheduling and insurance terms, in one transcript.
: Yes — Speech Understanding adds sentiment, topic detection, and PII/PHI redaction.
: AssemblyAI is a business associate under HIPAA and signs a standard BAA; PHI redaction; SOC 2 Type 2.
: Pay-as-you-go per second, no concurrency caps; Medical Mode $0.15/hr add-on.
: Add domain: "medical-v1" to the request.

Try Medical Mode free on your audio

Run your own patient-access call audio through Medical Mode in the playground — one parameter, real time. Free to start, no credit card required.

Try Medical Mode free