Corti launched Symphony for Speech-to-Text on May 20, 2026, a clinical-grade recognition model designed for high medical-terminology accuracy [3].
This development signals a shift toward specialized artificial intelligence in healthcare. By prioritizing clinical precision over the broad capabilities of general-purpose models, the technology aims to reduce documentation errors and administrative burdens for medical staff.
The Copenhagen-based company said Symphony achieves the highest medical-terminology accuracy recorded for this specific use case [2]. According to clinical accuracy benchmarks, the model outperforms OpenAI and other big-tech systems by over 25% [1].
Corti developed the tool to serve as a trusted AI scribe for physicians and patients. The company intends for the model to improve the reliability of clinical documentation, a critical component of patient safety and medical billing.
"We are focused on ensuring our AI scribes can be trusted by physicians, medical practitioners and patients...the entire healthcare system," Andreas Cleve said.
The launch highlights a growing divide between general large-language models and specialized AI. While big-tech models possess vast general knowledge, they often struggle with the nuanced, technical vocabulary required in a clinical setting. Symphony is designed to bridge that gap by focusing exclusively on medical coding and speech recognition.
“Symphony outperforms OpenAI and other big-tech models by over 25% in clinical accuracy benchmarks”
The emergence of clinical-grade models like Symphony suggests that 'vertical AI'—AI trained for a specific industry—may be more effective for high-stakes environments than general AI. In medicine, where a single terminology error can lead to diagnostic mistakes, the 25% accuracy gap over general models underscores the necessity of specialized training data over raw processing power.
