Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Multi-Source Evidence Fusion for Audio Question Answering

Aivo Olev, Tanel Alumäe

Abstract

Large audio language models (LALMs) can answer questions about speech, music, and environmental sounds, yet their internal reasoning is largely opaque and difficult to validate. We describe TalTech's solution to the Agent Track of the Interspeech 2026 Audio Reasoning Challenge, in which systems are evaluated on reasoning process quality, specifically the factual accuracy, logical soundness, and completeness of their reasoning chains. Our multi-source ensemble pipeline uses two LALMs that generate independent observations, while a separate text-only reasoning model cross-checks these against outputs from 25 acoustic tools organized into reliability tiers. By grounding every inference step in explicit, reliability-tagged evidence, the system produces dense, verifiable reasoning chains. Our system ranked first in the challenge, outperforming all competing systems by a wide margin in challenge's reasoning quality metric.

Multi-Source Evidence Fusion for Audio Question Answering

Abstract

Large audio language models (LALMs) can answer questions about speech, music, and environmental sounds, yet their internal reasoning is largely opaque and difficult to validate. We describe TalTech's solution to the Agent Track of the Interspeech 2026 Audio Reasoning Challenge, in which systems are evaluated on reasoning process quality, specifically the factual accuracy, logical soundness, and completeness of their reasoning chains. Our multi-source ensemble pipeline uses two LALMs that generate independent observations, while a separate text-only reasoning model cross-checks these against outputs from 25 acoustic tools organized into reliability tiers. By grounding every inference step in explicit, reliability-tagged evidence, the system produces dense, verifiable reasoning chains. Our system ranked first in the challenge, outperforming all competing systems by a wide margin in challenge's reasoning quality metric.
Paper Structure (13 sections, 1 figure, 6 tables)

This paper contains 13 sections, 1 figure, 6 tables.

Table of Contents

  1. Introduction
  2. Related work
  3. System architecture
  4. Experimental setup
  5. Dataset and metrics
  6. Implementation details
  7. Results
  8. Tool usage and usefulness
  9. Ablation: Multi-source evidence fusion
  10. Discussion
  11. Conclusion
  12. Acknowledgments
  13. Generative AI use disclosure

Figures (1)

  • Figure 1: Multi-source ensemble pipeline. Two speech LLMs analyze audio across four segments each; unified analysis identifies corroborations and disagreements. Step 1 gathers evidence from reliability-tiered tools (up to 3 rounds), three-stage contradiction detection generates verification hypotheses, and Step 2 performs targeted segment-level validation. Bracket counts indicate additional music-specific tools.