Generating Synthetic Doctor-Patient Conversations for Long-form Audio Summarization

arXiv:2604.06138 · cs.SD, cs.AI · Submitted 2026-04-07 · Read on arXiv

cs.SD, cs.AI

Submitted: 2026-04-07

Updated: 2026-10-11

Code: https://github.com/openai/whisper

Terminology

• Benchmark — 1× in this paper · explained in 2 episodes
• Synthetic Data Generation — 1× in this paper · explained in 5 episodes
• Synthetic Data Generation Pipeline — 1× in this paper · explained in 2 episodes

Sources

• Qwen3-Omni Technical Report
• BLAB: Brutally Long Audio Bench
• MMAU-Pro: A Challenging and Comprehensive Benchmark for Holistic Evaluation of Audio General Intelligence
• The Cascade Equivalence Hypothesis: When Do Speech LLMs Behave Like ASR$\rightarrow$LLM Pipelines?
• Gemma 3 Technical Report
• SDialog: A Python Toolkit for End-to-End Agent Building, User Simulation, Dialog Generation, and Evaluation
• Qwen3-ASR Technical Report
• Benchmarking Speech Systems for Frontline Health Conversations: The DISPLACE-M Challenge
• Qwen3 Technical Report
• Kimi K2: Open Agentic Intelligence

Related papers

• Few-Shot Open-Set Audio Classification via Transductive Prototype Refinement and Class Logit Enhancement
• Spectral Masking and Interpolation Attack (SMIA): A Black-box Adversarial Attack against Voice Authentication and Anti-Spoofing Systems
• Remix the Timbre: Diffusion-Based Style Transfer Across Polyphonic Stems
• AVMeme Exam: A Multimodal Multilingual Multicultural Benchmark for LLMs' Contextual and Cultural Knowledge and Thinking
• SoundWeaver: Compositional Warm-Starting for Text-to-Audio Diffusion Serving
• WASIL: In-the-Wild Arabic Spoken Interactions with LLMs