The paper introduces ToxiAlert-Bench, a large-scale audio dataset that uniquely annotates both textual and paralinguistic sources of toxicity, and proposes a dual-head neural network that significantl…

PolySpeech-100 introduces a massive, multi-lingual benchmark covering 110 linguistic variants to rigorously test Speech-LLMs, demonstrating that open-source models struggle with low-resource languages…

MindVoice is a neuro-to-speech framework that uses pretrained priors to disentangle and reconstruct intelligible speech from noisy, non-invasive neural signals, significantly outperforming existing me…

The paper introduces SURE, a unified framework designed to standardize and improve the comparability and reproducibility of evaluations for advanced speech understanding models.

ImmersiveTTS is an environment-aware text-to-speech model that generates natural speech seamlessly integrated within environmental contexts by explicitly modeling cross-modal interactions, achieving s…

HoliTok introduces a novel continuous holistic tokenization model that provides a unified, high-fidelity latent representation for simultaneously supporting both speech generation and speech understan…

The paper introduces three new Korean speech benchmarks (KVoiceBench, KOpenAudioBench, and KMMAU) to evaluate SpeechLMs, demonstrating that English-centric evaluation fails to capture performance gaps…

The paper proposes methods for generating global prosodic embeddings using auto-encoder models of pitch and energy, demonstrating competitive or superior performance under challenging conditions.

Murmur is an efficient inference system for long-form ASR that resolves the accuracy-latency trade-off by optimizing both inter-chunk processing and intra-chunk attention mechanisms.

MOSS-Audio is a unified audio-language model designed for comprehensive understanding of speech, environmental sounds, and music, achieving strong performance across various audio-grounded tasks.

The paper introduces OpenSTBench, a unified, multidimensional evaluation framework designed to comprehensively compare heterogeneous speech translation systems by jointly assessing translation, speech…

Sympatheia is a speech-to-speech dialogue framework that generates emotionally adaptive responses by conditioning its output on continuous affect signals derived from user speech or external multimoda…

A Wave-U-Net model is trained to extract a fundamental waveform from input speech signals for accurate and robust instantaneous pitch estimation.

This paper characterizes the gap between current DNN-based speech enhancement systems and hearing aid constraints, and proposes a lightweight architecture to meet these constraints.

SALSA is a lightweight adaptation method that learns layer-wise steering vectors to significantly improve the performance of speech-aware LLMs on out-of-domain speech tasks.

The paper proposes LaSR, a context-aware training paradigm that uses latent reasoning to significantly improve speech recognition, especially for specialized terminology, without adding latency.

The paper introduces an interpretable method for distinguishing genuine hate speech from contextually nuanced reclaimed language, achieving robust performance even with severe class imbalance.

UniAudio-Token is a framework that enhances existing semantic speech tokenizers with general audio perception, allowing them to handle diverse audio types while maintaining high-fidelity speech capabi…

The paper introduces a novel Clean-Referenced Feature-Vocoder Attack, a black-box adversarial attack that perturbs high-level SSL feature representations instead of raw audio waveforms, achieving supe…

This paper evaluates biases in multimodal speech recognition by testing how pairing different faces with the same audio affects transcription accuracy, finding significant quality-of-service drops acr…