TalTech systems win Beyond Transcription Challenge with novel speech-to-SOAP note generation

Researchers from Tallinn University of Technology (TalTech) have developed systems for the Beyond Transcription Challenge (BeTraC), which aims to generate SOAP notes directly from doctor-patient conversations without intermediate transcription. Their adapted models, Voxtral Mini and Voxtral Small, utilized LoRA fine-tuning and DAPO reinforcement learning with the Open Medical Concept F1 metric. These systems achieved first place in both lightweight and heavyweight tracks of the challenge, demonstrating a low hallucination rate and effective transfer of text-based fine-tuning to speech input. AI

IMPACT Demonstrates a novel approach to direct speech-to-SOAP note generation, potentially improving efficiency in clinical documentation.

RANK_REASON Academic paper detailing a novel approach to speech-to-text summarization for medical applications. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CL →

• Bétracq
• Beyond Transcription Challenge
• DAPO++
• Lora
• Open Medical Concept F1
• Tallinn University of Technology
• Voxtral Mini
• Voxtral Small

• paper
• product
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →