Searcharxiv⌕ Search
Home/Archive/arXiv

arXiv · 2610.40695

JYVA at BeTraC: Audio-Based Medical Note Generation with SOAP Section-Specific Queries

Youness Dkhissi·Valentin Vielzeuf·Antoine Caubrière·Jean-Emile Dartois
View original source ↗Read source document ↗Export metadata ↓

Abstract

We present our submission to the Beyond Transcription Challenge (BeTraC) on medical note generation from doctor-patient audio conversations. We propose a section-aware approach based on Qwen2.5-Omni-3B, in which a lightweight Q-Former learns SOAP section-specific queries to extract compact clinical representations from speech. In the lightweight track, our system ranks fourth and substantially outperforms the challenge baseline, achieving a Concept F1 of 0.5005 on the official test set. The organizers' analysis and our heuristic evaluation suggest improved robustness to hallucinations, and better stability under difficult audio conditions. These results show that explicitly modeling SOAP structure through section-specific queries is an effective way to improve the faithfulness and reliability of audio-based medical note generation.

Explore related subjects

cs.SDeess.AS

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Youness Dkhissi, Valentin Vielzeuf, Antoine Caubrière, Jean-Emile Dartois. 2026-10-28. JYVA at BeTraC: Audio-Based Medical Note Generation with SOAP Section-Specific Queries. https://arxiv.org/abs/2610.40695

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

Same Words, Different Judgments: How Preferences Vary Across Modalities

Preference-based reinforcement learning (PbRL) is the dominant framework for aligning AI systems to human preferences. However, evaluation protocols for such data were designed for text and have not been validated for speech. We present the first intraclass correlation coefficient (ICC)-based, controlled cross-modal study of human and synthetic preference annotations, comparing text and audio evaluations of identical semantic content across 100 prompts. We show that achieving \textit{good} agreement within either modality (ICC(2,$k$) $\approx$ .80) requires $\sim$9 raters and modalities show marked differences in how people report preferences: audio raters exhibit narrower decision thresholds, reduced length bias, and more user-oriented evaluation criteria, with near-chance cross-modality agreement. We demonstrate that synthetic ratings can be used to effectively predict inter-rater agreement, thus serving as an early signal for stimulus selection and proxy for human annotations. Together, these findings argue that evaluation protocols for audio preference data require modality-specific design rather than direct adaptation from text. Code, data, and the audio corpus are available at: https://huggingface.co/datasets/sciminds/same-words-different-judgements

cs.SD↗
arXiv2026

Oído: Real-Time Open-Vocabulary Speech Recognition on a $5 Microcontroller without a Neural Accelerator

Speech recognition on low-cost microcontrollers is, in practice, limited to keyword spotting or a fixed list of commands. We present Oído, an inference engine and a family of models that run open-vocabulary English speech recognition entirely on an ESP32-S3 (240 MHz Xtensa LX7, 8 MB of PSRAM, 16 MB of flash, no neural accelerator) faster than real time on one core. Version 1 of this work ran a 13 M-parameter Conformer-CTC at 1.97x the duration of the audio. We measured on the board that the bound was memory traffic and per-frame cost, not arithmetic: external memory reads at 89 MB/s (PSRAM) and 32 MB/s (flash), and a streaming recognizer re-reads every weight for every chunk. Version 2 combines a model built for that memory system (80 ms frame stacking, a 64-channel front end, 12 or 16 blocks, int4 weights that mostly fit in PSRAM, fine-tuned from the version 1 model) with engine changes that overlap the weight stream with compute (an asynchronous data-cache preload), keep work buffers in SRAM and make the decoder cheap. On a physical board the streaming real-time factor falls from 1.77 to 0.54 (variant B) and 0.72 (variant A), the final text appears 259 ms (B) or 431 ms (A) after the end of speech with audio fed at real-time speed, and the host build and the firmware produce identical transcripts. Variant A reaches 7.1%/16.4% word error rate on LibriSpeech test-clean/test-other in streaming mode with 0.64 s chunks and the on-chip language model (4.6%/10.7% with full context). Code, models and the raw board measurements are released.

cs.SD↗
arXiv2026

Biologically Inspired Predictive Gating Signed-Dyadic Coding for Mandarin Keyword Spotting

In continuous spiking keyword spotting, temporally correlated acoustic frames can repeatedly trigger memory access and neural computation at a dense input interface, even when hidden activity is sparse. Predictive processing and sensory adaptation suggest retaining predictable input locally and responding selectively to deviations. Inspired by this mechanism, we propose Predictive Gating Signed-Dyadic Coding (PG). PG maintains a local prediction for each input channel and transmits an event only when the residual exceeds a learned threshold. Each event carries a sign and a finite power-of-two magnitude, and the receiver reconstructs it with fixed-point shifts and additions. In Mandarin keyword spotting, INPUT-PG reaches 90.00% accuracy while reducing mean SRAM reads by 33.87% and inference cycles by 32.77%. PG provides a hardware interface for research on low-power continuous speech processing and event-based sensory systems.

cs.SD↗