Pith. sign in

REVIEW 7 cited by

Language Model Can Listen While Speaking

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2408.02622 v1 pith:CV5RPCQ6 submitted 2024-08-05 cs.CL cs.AIcs.HCcs.SDeess.AS

classification cs.CLcs.AIcs.HCcs.SDeess.AS
keywords fusionlslmspeechlanguagereal-timegenerationinteractionmodel
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Dialogue serves as the most natural manner of human-computer interaction (HCI). Recent advancements in speech language models (SLM) have significantly enhanced speech-based conversational AI. However, these models are limited to turn-based conversation, lacking the ability to interact with humans in real-time spoken scenarios, for example, being interrupted when the generated content is not satisfactory. To address these limitations, we explore full duplex modeling (FDM) in interactive speech language models (iSLM), focusing on enhancing real-time interaction and, more explicitly, exploring the quintessential ability of interruption. We introduce a novel model design, namely listening-while-speaking language model (LSLM), an end-to-end system equipped with both listening and speaking channels. Our LSLM employs a token-based decoder-only TTS for speech generation and a streaming self-supervised learning (SSL) encoder for real-time audio input. LSLM fuses both channels for autoregressive generation and detects turn-taking in real time. Three fusion strategies -- early fusion, middle fusion, and late fusion -- are explored, with middle fusion achieving an optimal balance between speech generation and real-time interaction. Two experimental settings, command-based FDM and voice-based FDM, demonstrate LSLM's robustness to noise and sensitivity to diverse instructions. Our results highlight LSLM's capability to achieve duplex communication with minimal impact on existing systems. This study aims to advance the development of interactive speech dialogue systems, enhancing their applicability in real-world contexts.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 7 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Open-Source Full-Duplex Conversational Datasets for Natural and Interactive Speech Synthesis

    cs.SD 2025-09 conditional novelty 6.0 of 10

    Two open-source dual-track conversational speech corpora (Chinese and English) are introduced and shown to modestly improve a fine-tuned TTS model's naturalness metrics.

  2. Towards Emotion Co-regulation with LLM-powered Socially Assistive Robots: Integrating LLM Prompts and Robotic Behaviors to Support Parent-Neurodivergent Child Dyads

    cs.HC 2025-07 conditional novelty 6.0 of 10

    An LLM-driven MiRo-E robot coached parents and neurodivergent children through stressful LEGO tasks, and a two-dyad qualitative pilot suggested it can support emotion co-regulation.

  3. NTPP: Generative Speech Language Modeling for Dual-Channel Spoken Dialogue via Next-Token-Pair Prediction

    cs.CL 2025-06 conditional novelty 6.0 of 10

    NTPP models dual-channel spoken dialogue by predicting both speakers' next speech tokens as a pair, achieving speaker-independent full-duplex generation in a decoder-only transformer.

  4. VoiceStar: Robust Zero-Shot Autoregressive TTS with Duration Control and Extrapolation

    eess.AS 2025-05 conditional novelty 6.0 of 10

    VoiceStar uses a progress-based rotary position embedding and mixed prompt training to give zero-shot voice cloning precise duration control and much longer output than training clips.

  5. SALM-Duplex: Efficient and Direct Duplex Modeling for Speech-to-Speech Language Model

    cs.CL 2025-05 conditional novelty 6.0 of 10

    A speech-to-speech language model uses channel fusion of a streaming encoder and codec tokens to handle barge-in and turn-taking without speech pretraining, showing improved metrics over Moshi at 0.6 kbps.

  6. Towards a Japanese Full-duplex Spoken Dialogue System

    cs.CL 2025-06 conditional novelty 5.0 of 10

    J-Moshi, the first public Japanese full-duplex spoken dialogue model, is built from Moshi and outperforms a Japanese dGSLM baseline on naturalness and meaningfulness.

  7. LiSTEN: Learning Soft Token Embeddings for Neural Audio LLMs

    cs.AI 2025-05 conditional novelty 5.0 of 10

    LiSTEN shows that dynamically selecting a few learnable prompt tokens from a shared pool can replace LoRA fine-tuning for audio-language models, matching or beating it with less training data.

Pith tools