Deepgram Launches Nova-3 Medical at HIMSS Conference
SAN FRANCISCO – March 3, 2025 – Deepgram, a leading voice AI platform for enterprise applications, today announced the launch of Nova-3 Medical. This next-generation, AI-powered speech-to-text (STT) model aims to meet the demanding requirements of the healthcare industry. The announcement was made at the HIMSS Conference in Las Vegas.
Nova-3 Medical is designed to enable developers to build highly accurate, customizable, and secure voice AI products and solutions tailored for healthcare settings. It seamlessly integrates with Deepgram’s enterprise runtime platform, including advanced text-to-speech (TTS) and speech-to-speech (STS) capabilities. This integration provides a comprehensive suite of AI-driven tools that deliver enterprise-grade performance, adaptability, and cost efficiency.
From streamlining clinical documentation to revolutionizing therapeutic scribing, Deepgram powers transformative medical transcription applications for industry leaders. The platform is engineered to drive exceptional outcomes across the healthcare spectrum.
Addressing the Rising Demand for Optimized Healthcare Transcription
As healthcare rapidly digitizes—with the widespread adoption of electronic health records, telemedicine, and digital health platforms—the demand for AI-powered transcription has surged. Traditional off-the-shelf speech-to-text models often struggle with the complexities of clinical terminology. This can lead to transcription errors and potentially problematic ‘hallucinations’ that may compromise patient care. The medical transcription market is projected to grow significantly, from USD 85.3 billion in 2023 to USD 190.2 billion by 2032, according to recent reports.
Developers building voice-AI applications for healthcare require infrastructure that provides exceptional speed and accuracy. They also need the flexibility to satisfy diverse regulatory and operational requirements.
Unveiling Nova-3 Medical’s Capabilities
Built to meet these demands, Nova-3 Medical uses advanced machine learning and specialized medical vocabulary training to set a new standard in healthcare transcription. The model accurately captures the unique terminology, acronyms, and jargon clinicians use, even in challenging audio conditions. It also delivers structured transcriptions that smoothly integrate with clinical workflows and EHR systems. This ensures vital patient data is accurately organized and readily accessible.
Its flexible, self-service customization features include Keyterm Prompting for up to 100 key terms. This feature allows developers to tailor the solution to the specific needs of particular medical specialties. Versatile deployment options, including on-premises and VPC configurations, ensure enterprise-grade security and HIPAA compliance.
“Nova-3 Medical represents a significant leap forward in our commitment to transforming clinical documentation through AI,” said Scott Stephenson, CEO of Deepgram. “By addressing the nuances of clinical language and offering unprecedented customization, we are empowering developers to build products that improve patient care and operational efficiency.”
Kevin Fredrick, Managing Partner at OneReach.ai, noted, “Speech-to-text for enterprise use cases is not trivial, and there is a fundamental difference between voice AI platforms designed for enterprise use cases vs. entertainment use cases.” He added, “Deepgram’s Nova-3 model and Nova-3 Medical model are leading voice AI offerings, including TTS, in terms of the inherent accuracy, latency, efficiency, and scalability required for enterprise use cases.”
Benchmarking Nova-3 Medical: Accuracy, Speed, and Efficiency
Nova-3 Medical delivers industry-leading transcription accuracy. It optimizes both overall word recognition and critical medical term accuracy for voice-driven healthcare applications.
Error Rate Comparison
With a median Word Error Rate (WER) of 3.45%, Nova-3 Medical outperforms competing models. This represents a 63.6% reduction in errors compared to the next best competitor. This increase in accuracy enhances documentation precision, leading to streamlined workflows for healthcare providers.
Critical Term Accuracy
Accurately capturing specialized medical terminology is essential for minimizing patient care risks. Nova-3 Medical achieves a Keyword Error Rate (KER) of 6.79%, marking a 40.35% reduction in errors compared to the next best competitor. This ensures that fewer critical drug names, conditions, and procedures are misrecognized, reducing the chances of transcription errors that could cause miscommunication, improper documentation, or even patient safety risks.
In addition to its accuracy, Nova-3 Medical is optimized for real-time performance, where speed and scalability are crucial. It transcribes speech 5 to 40 times faster than many alternative speech recognition vendors. This makes it well-suited for telemedicine and digital health platforms. Its scalable architecture ensures that healthcare tech companies can maintain high performance without incurring excessive costs as transcription volumes grow.
Starting at $0.0077 per minute of streaming audio, Nova-3 Medical is more than twice as affordable as leading cloud providers. This reduces operational expenses and enables companies to reinvest in innovation, accelerate product development, and offer competitive pricing to drive market adoption.
Deepgram at HIMSS25
Visit Deepgram at Booth #136 in the AI Pavilion at HIMSS25, March 3-6, 2025, to see Nova-3 Medical in action. They will also be hosting the following sessions:
- Session: From AI Scribes to EHR Automation: How Deepgram Enables Healthtech with Voice AI and Amazon Bedrock
- When: Tuesday, March 4, 3:40 PM to 4:00 PM
- Where: AI Pavilion, Venetian, Level 2, Hall A
- Session: Voice AI Mixer with Deepgram & OneReach.ai
- When: Wednesday, March 5, 6:00 PM to 7:30 PM
- Where: Venetian, Palazzo Ballroom, Palazzo A
For more information about Nova-3 Medical and how it is revolutionizing healthcare transcription, please visit www.deepgram.com.
About Deepgram
Deepgram is the leading voice AI platform for enterprise use cases, offering speech-to-text (STT), text-to-speech (TTS), and full speech-to-speech (STS) capabilities. Over 200,000 developers use their voice-native foundational models through cloud APIs or self-hosted / on-premises APIs because of the accuracy, low latency, and competitive pricing. Customers include technology ISVs, co-sell partners working with large enterprises, and enterprises solving internal use cases. Having processed over 50,000 years of audio and transcribed over 1 trillion words, Deepgram has extensive experience.
To learn more, visit www.deepgram.com, review their developer documentation, or follow @DeepgramAI on X and LinkedIn.