This paper introduces a training recipe for sentence-level and long-form streaming speech-to-speech translation using only 2k hours of paired cross-lingual data and auxiliary supervision.

ProactiveLLM introduces a novel framework that enables streaming LLMs to actively decide when to interact with incoming data by leveraging the model's internal states, significantly reducing latency w…

This paper proposes cumsum-composable phase transport, a streaming-native temporal layer for keyword spotting using unitary transport, prefix differences, and gated residual updates.

A lightweight framework for automated pronunciation assessment using native speech resources and unsupervised or lightly calibrated methods.

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…

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.

This paper proposes Joint Speech-Text Interleaved Pretraining (JSTIP) for speech recognition, which constructs interleaved speech-text sequences and achieves consistent entity accuracy improvement.

Chatterbox-Flash introduces a prior-calibrated block diffusion model for zero-shot TTS that achieves high-fidelity, streaming synthesis with significantly lower computational overhead than existing me…

This paper investigates timestamp drift in modern autoregressive ASR systems and proposes REDDIT, a two-stage post-training framework to correct timestamps while avoiding forgetting.

X-Translator is a low-cost modular system for real-time speech-to-speech translation, using streaming ASR, machine translation, and prompt-conditioned TTS, with session-level control.

This paper uses continual learning with explicit disfluency tokens to improve Automatic Speech Recognition (ASR) systems on disfluent speech, addressing the information loss and hallucinations caused…

CORTIS is a text-only adaptation framework that fine-tunes spoken language models for task-oriented voice agents using text-form task supervision.

This paper proposes SemBridge, a framework for continuous-latent autoregressive speech generation using discrete semantic tokens for supervision.