# VoxMem: Benchmarking Multimodal Memory in Large Audio Language Models

arXiv: https://arxiv.org/abs/2609.32607 · HTML: https://mlsanity.com/paper/2609.32607/

## Key facts

- No citing papers in the map yet (data through 19 Oct 2026).
- Placed in Audio language models and spoken dialogue, one of 11 directions in Speech and audio language models.
- Builds on 31 of its 73 references that are in the map.
- First posted on arXiv 26 Sep 2026.

## Placed in

- Speech and audio language models / [Audio language models and spoken dialogue](https://mlsanity.com/speech/audio-language-models-and-spoken-dialogue/)

## Builds on (most cited first, 30 of 31)

- [Evaluating Very Long-Term Conversational Memory of LLM Agents](https://mlsanity.com/paper/2402.17753/) (2024)
- [LongMemEval: Benchmarking Chat Assistants on Long-Term Interactive Memory](https://mlsanity.com/paper/2410.10813/) (2024)
- [Qwen-Audio: Advancing Universal Audio Understanding via Unified Large-Scale Audio-Language Models](https://mlsanity.com/paper/2311.07919/) (2023)
- [SALMONN: Towards Generic Hearing Abilities for Large Language Models](https://mlsanity.com/paper/2310.13289/) (2023)
- [Qwen3-Omni Technical Report](https://mlsanity.com/paper/2509.17765/) (2025)
- [Phi-4-Mini Technical Report: Compact yet Powerful Multimodal Language Models via Mixture-of-LoRAs](https://mlsanity.com/paper/2503.01743/) (2025)
- [LLaMA-Omni: Seamless Speech Interaction with Large Language Models](https://mlsanity.com/paper/2409.06666/) (2024)
- [Audio Flamingo 2: An Audio-Language Model with Long-Audio Understanding and Expert Reasoning Abilities](https://mlsanity.com/paper/2503.03983/) (2025)
- [Step-Audio 2 Technical Report](https://mlsanity.com/paper/2507.16632/) (2025)
- [Baichuan-Omni-1.5 Technical Report](https://mlsanity.com/paper/2501.15368/) (2025)
- [MiniCPM-o 4.5: Towards Real-Time Full-Duplex Omni-Modal Interaction](https://mlsanity.com/paper/2604.27393/) (2026)
- [Baichuan-Audio: A Unified Framework for End-to-End Speech Interaction](https://mlsanity.com/paper/2502.17239/) (2025)
- [SpokenWOZ: A Large-Scale Speech-Text Benchmark for Spoken Task-Oriented Dialogue Agents](https://mlsanity.com/paper/2305.13040/) (2023)
- [Conversation Chronicles: Towards Diverse Temporal and Relational Dynamics in Multi-Session Conversations](https://mlsanity.com/paper/2310.13420/) (2023)
- [AudioMarathon: A Comprehensive Benchmark for Long-Context Audio Understanding and Efficiency in Audio LLMs](https://mlsanity.com/paper/2510.07293/) (2025)
- [LongMemEval-V2: Evaluating Long-Term Agent Memory Toward Experienced Colleagues](https://mlsanity.com/paper/2605.12493/) (2026)
- [Audio Flamingo Next: Next-Generation Open Audio-Language Models for Speech, Sound, and Music](https://mlsanity.com/paper/2604.10905/) (2026)
- [SpeechPrune: Context-aware Token Pruning for Speech Information Retrieval](https://mlsanity.com/paper/2412.12009/) (2024)
- [Audio MultiChallenge: A Multi-Turn Evaluation of Spoken Dialogue Systems on Natural Human Interaction](https://mlsanity.com/paper/2512.14865/) (2025)
- [MTalk-Bench: Evaluating Speech-to-Speech Models in Multi-Turn Dialogues via Arena-style and Rubrics Protocols](https://mlsanity.com/paper/2508.18240/) (2025)
- [BLAB: Brutally Long Audio Bench](https://mlsanity.com/paper/2505.03054/) (2025)
- [MULTI-Bench: A Multi-Turn Interactive Benchmark for Assessing Emotional Intelligence ability of Spoken Dialogue Models](https://mlsanity.com/paper/2511.00850/) (2025)
- [Does Your Voice Assistant Remember? Analyzing Conversational Context Recall and Utilization in Voice Interaction Models](https://mlsanity.com/paper/2502.19759/) (2025)
- [LongSpeech: A Scalable Benchmark for Transcription, Translation and Understanding in Long Speech](https://mlsanity.com/paper/2601.13539/) (2026)
- [ChronosAudio: A Comprehensive Long-Audio Benchmark for Evaluating Audio-Large Language Models](https://mlsanity.com/paper/2601.04876/) (2026)
- [MemLens: Benchmarking Multimodal Long-Term Memory in Large Vision-Language Models](https://mlsanity.com/paper/2605.14906/) (2026)
- [FireRedAudio: A General-Purpose Audio Language Model with Decoupled Continuous Representations for Understanding and Generation](https://mlsanity.com/paper/2608.24168/) (2026)
- [Why Can't They Remember? Uncovering Representation and Retrieval Bottlenecks in Multi-Turn Acoustic Memory](https://mlsanity.com/paper/2605.27039/) (2026)
- [VoiceGiraffe: A Benchmark for Extreme Long-Context Audio-Language Understanding](https://mlsanity.com/paper/2605.27976/) (2026)
- [Vox-Infinity: Benchmarking the Limits of Long-Context Spoken Language Models](https://mlsanity.com/paper/2609.22452/) (2026)

## Abstract (arXiv)

Spoken conversational systems must recover information from prior interactions (i.e., memory), yet relevant information in speech extends beyond what was said to who said it, how it was spoken, and what was audible, information that exists only in the audio signal and cannot be recovered from a transcript. Beyond what to remember, memory also demands diverse operations: retrieving a single fact, integrating evidence across turns, tracking an evolving state. Real interactions further unfold across sessions, meaning information accumulates across distinct episodes rather than a single continuous recording. Existing benchmarks fall short on all three dimensions: they focus primarily on lexical content, adopt limited and ad hoc memory operations, and treat memory as a single-session problem. We argue that principled memory evaluation requires jointly characterizing the acoustic evidence to be retained and the operations applied to it, and introduce a taxonomy along these two axes. Building on this taxonomy, we present VoxMem: 3,196 evaluation instances over 34,743 spoken sessions (177 hours) crossing four acoustic evidence types (speech semantics, speaker identity, paralinguistic cues, environmental sound) with four memory operations (information extraction, multi-session reasoning, temporal tracking, and answer refusal), grounded in multi-session histories and stratified across context budgets from 8K to 64K tokens. Evaluating 15 LALMs, no model exceeds 40% at 32K. Models retain what was said far better than who said it, how, or what was audible, a gap that widens for complex operations, grows with history length, and manifests as qualitatively distinct failure modes across evidence types. VoxMem aims to provide a foundation to measure and drive progress on the full scope of spoken conversational memory.