VideoFDB is a new benchmark and LM-as-judge framework for evaluating full-duplex audio-visual-to-audio-visual conversational agents on nonverbal dynamics from real video calls.
hub Canonical reference
Moshi: a speech-text foundation model for real-time dialogue
Canonical reference. 73% of citing Pith papers cite this work as background.
68
Pith papers citing it
Background
73% of classified citations
abstract
We introduce Moshi, a speech-text foundation model and full-duplex spoken dialogue framework. Current systems for spoken dialogue rely on pipelines of independent components, namely voice activity detection, speech recognition, textual dialogue and text-to-speech. Such frameworks cannot emulate the experience of real conversations. First, their complexity induces a latency of several seconds between interactions. Second, text being the intermediate modality for dialogue, non-linguistic information that modifies meaning -- such as emotion or non-speech sounds -- is lost in the interaction. Finally, they rely on a segmentation into speaker turns, which does not take into account overlapping speech, interruptions and interjections. Moshi solves these independent issues altogether by casting spoken dialogue as speech-to-speech generation. Starting from a text language model backbone, Moshi generates speech as tokens from the residual quantizer of a neural audio codec, while modeling separately its own speech and that of the user into parallel streams. This allows for the removal of explicit speaker turns, and the modeling of arbitrary conversational dynamics. We moreover extend the hierarchical semantic-to-acoustic token generation of previous work to first predict time-aligned text tokens as a prefix to audio tokens. Not only this "Inner Monologue" method significantly improves the linguistic quality of generated speech, but we also illustrate how it can provide streaming speech recognition and text-to-speech. Our resulting model is the first real-time full-duplex spoken large language model, with a theoretical latency of 160ms, 200ms in practice, and is available at https://github.com/kyutai-labs/moshi.
hub tools
citation-role summary
background 12
baseline 3
citation-polarity summary
representative citing papers
Zero-Run auditing supplies valid lower bounds on differential privacy parameters from fixed member and non-member datasets by modeling and correcting distribution-shift confounding via causal-inference techniques.
PolySpeech-100 is a new benchmark for native-level speech comprehension across 110 linguistic variants that evaluates 22 models and reports E2E advantages on dialects, robustness gaps on low-resource languages, and degradation from Chain-of-Thought prompting.
PrefixMem encoder for Semantic IDs improves deepest-level accuracy by up to 46% relative and full-SID retrieval recall by up to 22% relative on Pinterest data across LLM families.
AffectCodec applies block-diagonal projections in residual FSQ to explicitly allocate bits to emotion and acoustic subspaces, combined with emotion conditioning, yielding better emotion preservation at low bitrates with competitive acoustic quality.
CodecAttack perturbs audio in codec latent space with multi-bitrate EoT to achieve 85.5% average ASR on Opus-compressed Audio LLMs versus under 26% for waveform baselines, with transfer to MP3 and AAC.
ClariCodec achieves 3.55% WER on LibriSpeech test-clean at 300 bps by RL fine-tuning the encoder for intelligibility, yielding a 23% relative WER reduction while preserving perceptual quality.
AffectCodec is an emotion-guided neural speech codec that preserves emotional cues during quantization while maintaining semantic fidelity and prosodic naturalness.
Channel fusion gives better semantic grounding and QA performance in full-duplex LLM dialogue but is vulnerable to context corruption during interruptions, while cross-attention routing is more robust at the cost of weaker integration.
VITA-QinYu is the first expressive end-to-end spoken language model supporting role-playing and singing alongside conversation, trained on 15.8K hours of data and outperforming prior models on expressiveness and conversational benchmarks.
LiVeAction is a lightweight asymmetric neural codec using an FFT-inspired encoder and variance-based training that outperforms generative tokenizers in rate-distortion while supporting real-time use on resource-constrained sensors across modalities.
Semantic priors from HuBERT and Whisper improve speech codec intelligibility up to 6 kbps but show diminishing returns beyond that, with a bitrate-aware regulation strategy balancing semantic consistency and naturalness.
SpeechParaling-Bench is a new evaluation framework for paralinguistic-aware speech generation that reveals major limitations in current large audio-language models.
Introduces the Indic-CodecFake dataset for Indic codec deepfakes and SATYAM, a novel hyperbolic ALM that outperforms baselines through dual-stage semantic-prosodic fusion using Bhattacharya distance.
AudioHijack generates imperceptible adversarial audio via gradient estimation, attention supervision, and reverberation blending to hijack 13 LALMs with 79-96% success on unseen contexts and real commercial agents.
HumDial-EIBench is a new benchmark using real human dialogues to evaluate audio language models on emotional intelligence tasks including multi-turn tracking, causal reasoning, empathy generation, and acoustic-semantic conflict resolution.
CapTalk unifies single-utterance and dialogue voice design via utterance- and speaker-level captions plus a hierarchical variational module for stable timbre with adaptive expression.
TiCo enables spoken dialogue models to follow explicit time constraints in generated responses using Spoken Time Markers and reinforcement learning with verifiable rewards, cutting duration error by 2.7x over its backbone.
Spoken language models exhibit style amnesia and fail to maintain instructed paralinguistic styles across multi-turn conversations, with explicit recall offering partial mitigation.
ViBES introduces a speech-language-behavior model using modality-specific transformer experts that jointly generates dialogue and 3D body actions, showing gains over separate co-speech and text-to-motion baselines on multi-turn metrics.
Game-Time Benchmark shows spoken language models handle basic tasks but degrade sharply under temporal constraints like tempo adherence and synchronized responses.
VoiceBench is the first benchmark for multi-faceted evaluation of LLM voice assistants using real and synthetic spoken instructions with speaker, environmental, and content variations.
A multi-axis RL alignment technique improves pause handling, turn-taking, backchanneling, and interruption response in full-duplex spoken dialogue models by optimizing axis-specific rewards derived from human audio segments.
CleanCodec reframes audio tokenization as a selective information bottleneck to encode only perceptually important features at 12.5 tokens per second, outperforming prior codecs in efficiency, speaker similarity, and intelligibility.
citing papers explorer
-
VideoFDB: Evaluating Full-Duplex Vision-Speech Capabilities in Conversational Agents
VideoFDB is a new benchmark and LM-as-judge framework for evaluating full-duplex audio-visual-to-audio-visual conversational agents on nonverbal dynamics from real video calls.
-
Privacy Auditing with Zero (0) Training Run
Zero-Run auditing supplies valid lower bounds on differential privacy parameters from fixed member and non-member datasets by modeling and correcting distribution-shift confounding via causal-inference techniques.
-
PolySpeech-100: A Large-Scale Benchmark for Speech Understanding Across 100+ Languages and Dialects
PolySpeech-100 is a new benchmark for native-level speech comprehension across 110 linguistic variants that evaluates 22 models and reports E2E advantages on dialects, robustness gaps on low-resource languages, and degradation from Chain-of-Thought prompting.
-
LLMs Need Encoders for Semantic IDs Too
PrefixMem encoder for Semantic IDs improves deepest-level accuracy by up to 46% relative and full-SID retrieval recall by up to 22% relative on Pinterest data across LLM families.
-
AffectCodec: Emotion-Preserving Neural Speech Codec with Block-Diagonal Residual FSQ
AffectCodec applies block-diagonal projections in residual FSQ to explicitly allocate bits to emotion and acoustic subspaces, combined with emotion conditioning, yielding better emotion preservation at low bitrates with competitive acoustic quality.
-
Codec-Robust Attacks on Audio LLMs
CodecAttack perturbs audio in codec latent space with multi-bitrate EoT to achieve 85.5% average ASR on Opus-compressed Audio LLMs versus under 26% for waveform baselines, with transfer to MP3 and AAC.
-
Optimising Neural Speech Codecs for 300bps Communication using Reinforcement Learning
ClariCodec achieves 3.55% WER on LibriSpeech test-clean at 300 bps by RL fine-tuning the encoder for intelligibility, yielding a 23% relative WER reduction while preserving perceptual quality.
-
AffectCodec: Emotion-Preserving Neural Speech Codec for Expressive Speech Modeling
AffectCodec is an emotion-guided neural speech codec that preserves emotional cues during quantization while maintaining semantic fidelity and prosodic naturalness.
-
How Should LLMs Listen While Speaking? A Study of User-Stream Routing in Full-Duplex Spoken Dialogue
Channel fusion gives better semantic grounding and QA performance in full-duplex LLM dialogue but is vulnerable to context corruption during interruptions, while cross-attention routing is more robust at the cost of weaker integration.
-
VITA-QinYu: Expressive Spoken Language Model for Role-Playing and Singing
VITA-QinYu is the first expressive end-to-end spoken language model supporting role-playing and singing alongside conversation, trained on 15.8K hours of data and outperforming prior models on expressiveness and conversational benchmarks.
-
LiVeAction: a Lightweight, Versatile, and Asymmetric Neural Codec Design for Real-time Operation
LiVeAction is a lightweight asymmetric neural codec using an FFT-inspired encoder and variance-based training that outperforms generative tokenizers in rate-distortion while supporting real-time use on resource-constrained sensors across modalities.
-
SPG-Codec: Exploring the Role and Boundaries of Semantic Priors in Ultra-Low-Bitrate Neural Speech Coding
Semantic priors from HuBERT and Whisper improve speech codec intelligibility up to 6 kbps but show diminishing returns beyond that, with a bitrate-aware regulation strategy balancing semantic consistency and naturalness.
-
SpeechParaling-Bench: A Comprehensive Benchmark for Paralinguistic-Aware Speech Generation
SpeechParaling-Bench is a new evaluation framework for paralinguistic-aware speech generation that reveals major limitations in current large audio-language models.
-
Indic-CodecFake meets SATYAM: Towards Detecting Neural Audio Codec Synthesized Speech Deepfakes in Indic Languages
Introduces the Indic-CodecFake dataset for Indic codec deepfakes and SATYAM, a novel hyperbolic ALM that outperforms baselines through dual-stage semantic-prosodic fusion using Bhattacharya distance.
-
Hijacking Large Audio-Language Models via Context-Agnostic and Imperceptible Auditory Prompt Injection
AudioHijack generates imperceptible adversarial audio via gradient estimation, attention supervision, and reverberation blending to hijack 13 LALMs with 79-96% success on unseen contexts and real commercial agents.
-
HumDial-EIBench: A Human-Recorded Multi-Turn Emotional Intelligence Benchmark for Audio Language Models
HumDial-EIBench is a new benchmark using real human dialogues to evaluate audio language models on emotional intelligence tasks including multi-turn tracking, causal reasoning, empathy generation, and acoustic-semantic conflict resolution.
-
CapTalk: Unified Voice Design for Single-Utterance and Dialogue Speech Generation
CapTalk unifies single-utterance and dialogue voice design via utterance- and speaker-level captions plus a hierarchical variational module for stable timbre with adaptive expression.
-
TiCo: Time-Controllable Spoken Dialogue Model
TiCo enables spoken dialogue models to follow explicit time constraints in generated responses using Spoken Time Markers and reinforcement learning with verifiable rewards, cutting duration error by 2.7x over its backbone.
-
Style Amnesia: Investigating Speaking Style Degradation and Mitigation in Multi-Turn Spoken Language Models
Spoken language models exhibit style amnesia and fail to maintain instructed paralinguistic styles across multi-turn conversations, with explicit recall offering partial mitigation.
-
ViBES: A Conversational Agent with Behaviorally-Intelligent 3D Virtual Body
ViBES introduces a speech-language-behavior model using modality-specific transformer experts that jointly generates dialogue and 3D body actions, showing gains over separate co-speech and text-to-motion baselines on multi-turn metrics.
-
Game-Time: Evaluating Temporal Dynamics in Spoken Language Models
Game-Time Benchmark shows spoken language models handle basic tasks but degrade sharply under temporal constraints like tempo adherence and synchronized responses.
-
VoiceBench: Benchmarking LLM-Based Voice Assistants
VoiceBench is the first benchmark for multi-faceted evaluation of LLM voice assistants using real and synthetic spoken instructions with speaker, environmental, and content variations.
-
Multi-Faceted Interactivity Alignment in Full-Duplex Speech Models
A multi-axis RL alignment technique improves pause handling, turn-taking, backchanneling, and interruption response in full-duplex spoken dialogue models by optimizing axis-specific rewards derived from human audio segments.
-
CleanCodec: Efficient and Robust Speech Tokenization via Perceptually Guided Encoding
CleanCodec reframes audio tokenization as a selective information bottleneck to encode only perceptually important features at 12.5 tokens per second, outperforming prior codecs in efficiency, speaker similarity, and intelligibility.
-
Taming Audio VAEs via Target-KL Regularization
The paper introduces target-KL regularization to train audio VAEs at specific bitrates, enabling rate-distortion curves and comparison to discrete audio codecs for improved text-to-sound generation.
-
Break-the-Beat! Controllable MIDI-to-Drum Audio Synthesis
Break-the-Beat! renders drum MIDI audio that matches the timbre of a reference clip by fine-tuning a text-to-audio model with a content encoder and hybrid conditioning on a new paired dataset.
-
Exploring Token-Space Manipulation in Latent Audio Tokenizers
LATTE creates a compact latent token bottleneck in audio tokenizers that aggregates global information and enables unsupervised editing of attributes like speaker identity via token swapping.
-
PoDAR: Power-Disentangled Audio Representation for Generative Modeling
PoDAR disentangles audio signal power from semantic content in latents using power augmentation and consistency objectives, yielding 2x faster convergence and gains of 0.055 speaker similarity and 0.22 UTMOS when applied to Stable Audio VAE with F5-TTS.
-
Reducing Linguistic Hallucination in LM-Based Speech Enhancement via Noise-Invariant Acoustic-Semantic Distillation
L3-SE reduces linguistic hallucination in LM-based speech enhancement by distilling noise-invariant acoustic-semantic representations from noisy inputs to condition an autoregressive decoder-only language model.
-
Continuous First, Discrete Later: VQ-VAEs Without Dimensional Collapse
An initial continuous autoencoder training phase prevents dimensional collapse in VQ-VAEs and yields lower reconstruction and perceptual losses.
-
VocalParse: Towards Unified and Scalable Singing Voice Transcription with Large Audio Language Models
VocalParse applies interleaved and Chain-of-Thought prompting to a Large Audio Language Model to jointly transcribe lyrics, melody and word-note alignments, achieving state-of-the-art results on multiple singing datasets.
-
MiniMind-O Technical Report: An Open Small-Scale Speech-Native Omni Model
MiniMind-O delivers a working 0.1B-scale open omni model with speech-native output, Thinker-Talker split, frozen encoders, and full release of code, checkpoints, and training data.
-
Aligning Backchannel and Dialogue Context Representations via Contrastive LLM Fine-Tuning
A contrastive LLM fine-tuning method creates joint embeddings for dialogue contexts and backchannel realizations, improving retrieval performance and alignment with human judgments over raw WavLM features.
-
Why Your Tokenizer Fails in Information Fusion: A Timing-Aware Pre-Quantization Fusion for Video-Enhanced Audio Tokenization
A timing-aware pre-quantization fusion approach integrates visual cues into audio tokenizers along the temporal axis, maintaining reconstruction quality while outperforming audio-only and prior multimodal baselines on downstream tasks.
-
Bridging What the Model Thinks and How It Speaks: Self-Aware Speech Language Models for Expressive Speech Generation
SA-SLM uses variational information bottleneck for intent-aware bridging and self-criticism for realization-aware alignment to close the semantic-acoustic gap, outperforming open-source models and nearing GPT-4o-Audio expressiveness on EchoMind after training on 800 hours of data.
-
ASPIRin: Action Space Projection for Interactivity-Optimized Reinforcement Learning in Full-Duplex Speech Language Models
ASPIRin decouples speaking timing from token content via binary action space projection and applies GRPO with rule-based rewards to optimize interactivity in SLMs without semantic collapse or repetition.
-
PoM: A Linear-Time Replacement for Attention with the Polynomial Mixer
PoM is a new linear-complexity token mixer using learned polynomials that matches attention performance in transformers while enabling efficient long-sequence processing.
-
FastTurn: Unifying Acoustic and Streaming Semantic Cues for Low-Latency and Robust Turn Detection
FastTurn unifies acoustic features and streaming CTC decoding for low-latency, robust turn detection in full-duplex dialogue systems and releases a realistic human-dialogue test set.
-
Qwen3-TTS Technical Report
Qwen3-TTS delivers state-of-the-art multilingual TTS performance with 3-second voice cloning, description control, and ultra-low-latency streaming via dual tokenizers and a dual-track LM architecture trained on over 5 million hours of data.
-
Asynchronous Reasoning: Training-Free Interactive Thinking LLMs
Using properties of positional embeddings, reasoning LLMs can be made to think, listen, and generate outputs asynchronously without any additional training, cutting time to first token to under 5 seconds.
-
Mind-Paced Speaking: A Dual-Brain Approach to Real-Time Reasoning in Spoken Language Models
MPS proposes a dual-brain architecture separating formulation reasoning from articulation to achieve real-time CoT in SLMs with accuracy comparable to full pre-computation but much lower latency.
-
StableToken: A Noise-Robust Semantic Speech Tokenizer for Resilient SpeechLLMs
StableToken introduces a multi-branch architecture with bit-wise voting to create noise-robust semantic speech tokens, achieving lower Unit Edit Distance and better SpeechLLM robustness than prior single-path tokenizers.
-
Training-Free Multimodal Large Language Model Orchestration
LLM Orchestration integrates modality experts via an LLM controller, cross-modal memory, and interaction layer to enable multimodal input-output without gradient-based training.
-
Step-Audio 2 Technical Report
Step-Audio 2 integrates a latent audio encoder, reasoning-centric reinforcement learning, and discrete audio token generation into language modeling to deliver state-of-the-art performance on audio understanding and conversational benchmarks.
-
ZipVoice-Dialog: Non-Autoregressive Spoken Dialogue Generation with Flow Matching
ZipVoice-Dialog is a flow-matching non-autoregressive model for zero-shot spoken dialogue generation that uses curriculum learning and speaker-turn embeddings, paired with a new 6.8k-hour OpenDialog dataset, and reports better speed and quality than autoregressive baselines.
-
SmolVLA: A Vision-Language-Action Model for Affordable and Efficient Robotics
SmolVLA is a small efficient VLA model that achieves performance comparable to 10x larger models while training on one GPU and deploying on consumer hardware via community data and chunked asynchronous action prediction.
-
Step-Audio: Unified Understanding and Generation in Intelligent Speech Interaction
Step-Audio introduces a 130B-parameter unified speech-text model with open-sourced components for understanding, generation, affordable voice cloning, and dynamic control, claiming SOTA human evaluation results on a new benchmark.
-
Sympatheia: Emotionally Adaptive Voice Assistant with Continuous Affect Conditioning
Sympatheia introduces a continuous affect-conditioned speech dialogue model and the Sympatheia-18k synthetic dataset, showing improved emotional appropriateness over baselines when speech cues are limited.
-
Synchronization and Turn-Taking in Full-Duplex Speech Dialogue Models
Full-duplex SDMs show strong representational synchronization that peaks near zero lag and degrades with noise, with internal states encoding anticipatory turn-taking cues detectable ahead of time.
-
A Survey of Large Audio Language Models: Generalization, Trustworthiness, and Outlook
A survey of Large Audio Language Models that establishes a taxonomy of trustworthiness vulnerabilities and proposes a Defense-in-Depth roadmap for audio intelligence.