AIAny
Icon for item

VoiceMem: Streaming Dual-Brain Memory for Real-Time Interaction

Provides a streaming dual-brain memory for real-time speech agents: an informational left brain for factual retrieval and an affective right brain for persona/emotion, achieving high top-5 accuracy while keeping retrieval latency within VAD budgets (~134 ms).

Introduction

Real-time conversational speech systems need memories that both stream as the user speaks and capture emotional and persona cues — not just facts. VoiceMem answers this by splitting memory into two cooperating tracks: a schema-driven left brain for precise fact retrieval and a right brain that stores affective and persona signals, all wired into a streaming ingestion and retrieval pipeline designed for sub-VAD latency.

Key Findings
  • Accuracy: The left-brain schema/entity design yields much higher effective retrieval under a Top-5 injection budget, outperforming classical wide-index approaches while keeping injected context small. This design reduces expensive context bloat during generation.
  • Emotional & Personal Modeling: The right brain uses short- and long-horizon affect attribution plus dual-node persona modeling to capture who the user is and how they feel, improving persona benchmarks and aggregate scores versus prior systems.
  • Real-time & Cost: A streaming I/O and hierarchical storage strategy (compression, layered retrieval, speculative prefetch) yields median retrieval times reported around 134 ms and keeps injected memory tokens low, enabling memory-aware SLMs without adding perceptible conversational delay.
Who it's for and trade-offs

Great fit if you build or evaluate real-time speech agents (duplex SLMs, voice assistants, on-device agents) that must personalize across turns while respecting strict latency budgets. It helps when you need both accurate factual recall and ongoing emotion/persona modeling. Look elsewhere if you only need simple session-scoped context or if you cannot afford the extra pipeline components (ASR, voice fingerprinting, affect extractors) or have strict privacy regimes that forbid storing user-sensitive values without additional safeguards.

How it works (brief)

VoiceMem ingests audio streams, runs ASR and audio-native perception (speaker ID, emotion, scene), extracts embeddings, and writes structured nodes into two cooperatively queried stores. Queries are routed and ranked, then only the Top-K memories are injected into the model context. The architecture is backend-agnostic so retrieval DBs and perception modules can be swapped per deployment needs.

Information

  • Websitearxiv.org
  • OrganizationsNanyang Technological University, National University of Singapore, Tsinghua University, The Chinese University of Hong Kong, Open Interaction Lab
  • AuthorsZhifei Xie, Jiaqi Lang, Ze An, Yifan Zhao, Dongchao Yang, Kai Li, Ziyang Ma, Mingbao Lin, Chunyan Miao, Shuicheng Yan
  • Published date2026/08/26

More Items

Supervises audio reasoning by generating per-sample, audio-grounded rubrics that evolve with model rollouts and serve as reinforcement-learning rewards, improving perception and adaptive multi-step reasoning while avoiding reward saturation.

Retrieves short speech segments from MEG recordings with a compact interpretable neural decoder trained against wav2vec 2.0 embeddings, and maps decoder weights to cortical source space to reveal which acoustic and linguistic features drive retrieval.

Generates multi‑speaker speech and environmental audio from textual instructions or a reference clip, supporting zero‑shot voice cloning and detailed scene/specification control. Combines a cleaned, captioned dataset with a VAE-based multimodal generator, reward-conditioned quality control, and staged training to improve expressiveness and multi-audio modeling.