https://speech-expert.ru/news/kanadapter-efficient-finetuning-foundation-speech-models Speech Expert ru 2026-09-07T00:00:00Z KanAdapter: A Kolmogorov-Arnold Network-based Plug-and-Play Module for Efficient Fine-tuning of Foundation Speech Models https://speech-expert.ru/news/enhancing-neural-speech-coding-semantic-visual-cues Speech Expert ru 2026-09-07T00:00:00Z Enhancing Neural Speech Coding with Semantic and Visual Cues https://speech-expert.ru/news/prolombard-normal-to-lombard-speech-conversion Speech Expert ru 2026-09-07T00:00:00Z ProLombard: Structured Multi-Scale Modeling for Normal-to-Lombard Speech Conversion https://speech-expert.ru/news/what-selects-reconstructs-exemplar-spectrum-separation Speech Expert ru 2026-09-07T00:00:00Z What Selects, What Reconstructs: Repairing Exemplar-Based Complex-Spectrum Separation https://speech-expert.ru/news/tracing-audio-grounding-answer-selection-audio-llms Speech Expert ru 2026-09-07T00:00:00Z Tracing Audio Grounding and Answer Selection in Audio LLMs https://speech-expert.ru/news/discriminative-flow-matching-speech-restoration Speech Expert ru 2026-09-07T00:00:00Z Discriminative Flow Matching: Beyond Time-Conditioning in Generative Restoration via Flow-State Representations https://speech-expert.ru/news/brain2speech-net-real-time-synthesis-without-text Speech Expert ru 2026-09-07T00:00:00Z Brain2Speech-Net: Intelligible, Real-Time Brain-to-Speech Synthesis Without Text Decoding https://speech-expert.ru/news/anatomy-of-an-asr-hallucination Speech Expert ru 2026-09-07T00:00:00Z The Anatomy of an ASR Hallucination https://speech-expert.ru/news/scalable-context-orchestration-serving-llms-voice Speech Expert ru 2026-09-07T00:00:00Z Scalable Context Orchestration for Serving LLMs Over Voice https://speech-expert.ru/news/ghostword-fine-grained-backdoor-asr Speech Expert ru 2026-09-07T00:00:00Z GhostWord: A Fine-Grained Backdoor Attack on Automatic Speech Recognition https://speech-expert.ru/news/cad-conflict-aware-decoding-cross-modal-hallucinations Speech Expert ru 2026-09-07T00:00:00Z CAD: Conflict-Aware Decoding to Mitigate Cross-Modal Hallucinations in Omnimodal Large Language Models https://speech-expert.ru/news/grounded-decoding-autoregressive-speech-enhancement Speech Expert ru 2026-09-07T00:00:00Z Grounded Decoding for Autoregressive Speech Enhancement via Adaptive Code-Space Grounding and Local LLM Refinement https://speech-expert.ru/news/turnfsm-full-duplex-dialogue-state-machine Speech Expert ru 2026-09-07T00:00:00Z TurnFSM for Full-Duplex Dialogue System: Internalizing State-Machine Logic for Streaming Semantic Voice Activity Detection and Utterance-Level Rejection https://speech-expert.ru/news/rethinking-speech-codecs-autoregressive-modeling Speech Expert ru 2026-09-07T00:00:00Z Rethinking Speech Codecs: From Compression to Autoregressive Generative Modeling https://speech-expert.ru/news/effvoc-low-delay-waveform-reconstruction-without-phase Speech Expert ru 2026-09-07T00:00:00Z EffVOC: Low-Delay Efficient Speech Waveform Reconstruction from Spectral Representations Without Phase https://speech-expert.ru/news/trade-off-labels-causal-turn-aware-streaming-asr Speech Expert ru 2026-09-07T00:00:00Z The Trade-off Was in the Labels: Causal Supervision for Turn-Aware Streaming ASR https://speech-expert.ru/news/gepard-generative-prosody-autoregressive-realtime-tts Speech Expert ru 2026-09-07T00:00:00Z GEPARD - Generative, Prosody-aware, Autoregressive text-to-speech model for Realtime Dialogue https://speech-expert.ru/news/compressing-streaming-audio-encoders-latent-distillation Speech Expert ru 2026-09-07T00:00:00Z Compressing Streaming Neural Audio Encoders via Latent-Space Distillation https://speech-expert.ru/news/deep-neural-compression-rir-acoustic-environments Speech Expert ru 2026-09-07T00:00:00Z Deep Neural Compression for RIR-Characterized Acoustic Environments with Structure-Aware Constraints https://speech-expert.ru/news/alignment-free-text-audiobox-voice-dubbing-dialogue Speech Expert ru 2026-09-07T00:00:00Z Alignment-Free Text-Audiobox for Voice Dubbing and Full-Duplex Dialogue Synthesis https://speech-expert.ru/news/masked-autoregressive-speech-enhancement-continuous-codec Speech Expert ru 2026-09-07T00:00:00Z Masked Autoregressive Speech Enhancement with Continuous Neural Audio Codec Representations https://speech-expert.ru/news/test-time-adaptation-speech-enhancement-autoregressive-prior Speech Expert ru 2026-09-07T00:00:00Z Test-time adaptation for speech enhancement with an autoregressive speech prior https://speech-expert.ru/news/broadband-acoustic-intensity-tight-frame-cardioid-arrays Speech Expert ru 2026-09-07T00:00:00Z Broadband Acoustic Intensity Direction Estimation with Tight-Frame Cardioid Arrays https://speech-expert.ru/news/geometric-ceilings-time-frequency-masking Speech Expert ru 2026-09-07T00:00:00Z Geometric Ceilings on Time-Frequency Masking for Single-Channel Separation https://speech-expert.ru/news/streamwsr-streamable-waveform-speech-super-resolution Speech Expert ru 2026-09-07T00:00:00Z StreamWSR: Streamable and Lightweight Waveform-Domain Neural Speech Super-Resolution https://speech-expert.ru/news/semantics-enough-speech-mos-prediction Speech Expert ru 2026-09-07T00:00:00Z Is Semantics Enough for Speech Mean Opinion Score Prediction? https://speech-expert.ru/news/listen-latents-self-correcting-speech-recognition Speech Expert ru 2026-09-07T00:00:00Z Listen to the Latents: Self-Correcting Speech Recognition in Large Audio Language Models Through Hidden-State Interactions https://speech-expert.ru/news/idsd-iterative-deep-learning-signal-decomposition Speech Expert ru 2026-09-07T00:00:00Z IDSD: Iterative Deep-Learning-Based Signal Decomposition