customers
All customer stories
Top Voice AI companies are building with Assembly.
resources
Latest Release
Universal 3.5 Pro
Our new flagship speech-to-text model handles real-world audio the way it actually happens. Available for real-time and pre-recorded audio.
resources
Process recorded encounters, dictation, and telehealth sessions with clinical-grade accuracy — optimized for medical terminology, speaker separation, and EHR-ready output.
10X
customer growth
“The new Universal-3.5 Pro speech model from AssemblyAI is best so far in terms of accuracy, latency, and language switching.”
80%
increase in customer satisfaction
“Assembly has saved us countless hours managing models, and provided exceptional accuracy.”
75%
engineering time savings on infrastructure
10X
customer growth
“The new Universal-3.5 Pro speech model from AssemblyAI is best so far in terms of accuracy, latency, and language switching.”
80%
increase in customer satisfaction
“Assembly has saved us countless hours managing models, and provided exceptional accuracy.”
75%
engineering time savings on infrastructure
Transform recorded clinical audio into accurate, structured medical documentation.
Activate Medical Mode for pharma names, dosages, and procedures
Deliver clinical-grade precision for diagnostic language
Recognize ICD/CPT terms out of the box
Redact PHI across text and audio, backed by a signed BAA
Maintain SOC 2 Type II certification with configurable data retention
BAA available for customers processing PHI under HIPAA
Accurately separate physician, patient, and staff speech
Attribute roles for structured clinical notes
Support multi-party encounters with complex speaker dynamics
Clinical-grade accuracy on recorded audio, with Medical Mode reducing medical entity errors by up to 87%.
Clinical-grade streaming accuracy
Highest-accuracy batch transcription for recorded clinical encounters, physician dictation, telehealth recordings, and surgical documentation with Medical Mode for clinical terminology.
Sub-300ms real-time accuracy for live telehealth transcription and real-time clinical documentation capture.
Instant clinical voice input — one HTTP call returns a flagship-accuracy transcript in ~134 ms for push-to-talk dictation and short voice notes up to 2 minutes.
The terms that determine patient outcomes — medication names, dosages, and diagnoses — transcribed more accurately than ever.
| AssemblyAI Universal-3.5 Pro w/ Medical Mode | Speechmatics Enhanced Medical | Deepgram Nova-3 Medical | AWS Transcribe Medical | Google Medical Conversation | |
|---|---|---|---|---|---|
| MER | 3.2% | 3.6% | 4.7% | 8.7% | 24.4% |
| WER | 5.3% | 5.5% | 6.1% | 5.9% | 12.9% |
Put our Voice AI models to the test in our no-code playground.
Try it now
