Engineering & Signals · Interdisciplinary & Society
Audio & Speech Processing
The newest papers in Audio & Speech Processing that ReadSearch has read, newest first. Each links to its own page, with its authors, date, where it was published and the original.
334 papers, from 8 September 2026 to 4 November 2026: 334 arXiv preprints.
Get the few papers worth your time in Audio & Speech Processing every morning.
Follow Audio & Speech Processing and ReadSearch reads its new papers each day, then brings you the few worth reading, each with its key finding in a sentence.
Newest papers, page 6
Repairing Speech Access After Text Updates with Frozen Affine Fields
Speech-capable language models inherit most of their knowledge from text-rich language backbones, and continued model maintenance is likewise dominated by text updates.SNOAR: Self-training with Noise filtering, Observation adding, and Remixing for Speech Denoising Without Clean References
Unsupervised speech denoising methods such as MixIT and noisy-target training can be trained directly on noisy recordings, avoiding the sim-to-real gap of synthetic training data, but remain well…MIDI-REPA: Improving Text-to-Music Generation with Symbolic Representation Alignment
State-of-the-art text-to-music models increasingly rely on scaling model capacity, data, and compute, while still learning musical structure directly from dense audio.From Individual Songs to Music Corpora: Evaluating Explainable Harmonic Decision-Making on HookTheory
Automatic melody harmonization is often evaluated through individual compositions or data-driven corpus models.GeoMER: A Joint Euclidean-Hyperbolic Geometry Framework for Multimodal Emotion Recognition
Multimodal Emotion Recognition (MER) aims to understand human emotions by integrating information from multiple modalities such as speech and text.Deep Speech Representations Track Word Timing along a Low-Dimensional Circular Trajectory
Speech timing is fundamental for speech models to recognize words accurately, but how word-timing information is represented in speech models remains underexplored.D2: High-Capability Full-Duplex Spoken Language Models from Half-Duplex Backbones
Full-duplex spoken language models (FD-SLMs) support simultaneous listening and speaking, but often lag behind half-duplex spoken language models (HD-SLMs) in general capabilities.SonicWorld: Source-Centric Latent Dynamics for Spatial Audio Reasoning
Dynamic spatial audio reasoning requires a model to associate changing acoustic observations with the correct source and recover its spatial state at a queried time or interval, a challenge…CIViC: Evidence Gates for Audio-Conditioned Visual Prediction on Proxy Tasks
Removing audio can shift a visual predictor's output without showing whether the task uses the intended audio information.JSIR: An Audio-Aligned, Multi-Task Dataset of Jazz Standard Recordings
Beat tracking, chord estimation, and structure analysis are staple tasks of music information retrieval.LiveAnon: Improving Privacy and Utility in Streaming Speaker Anonymization with Codec Language Models
Speaker anonymization (SA) aims to conceal speaker identity while preserving linguistic content and emotion, but existing streaming systems often lose emotion information through discrete content…BridgeCodec: Mamba Enhanced Neural Audio Codec with Schrödinger Bridge at Low Bitrate
Neural audio codecs achieve remarkable performance but operate as closed ecosystems with rigidly co-trained encoderdecoder pairs, severely hindering interoperability.Explore This! Beat and Downbeat Tracking From a Learned Tatum Grid
State-of-the-art beat tracking is based on neural networks that model beats and downbeats as individual events, often followed by a separate Dynamic Bayesian Network (DBN) enforcing musical…Cross-Lingual TTS for Indian Languages: A Benchmark and Inference-Time Duration Estimation Strategies
Recent alignment-free Text-to-Speech (TTS) models achieve high synthesis quality by removing explicit alignment modules and simplifying training.Word-Aligned Prosody Reflects Speaker Differences More Than Depression or PTSD Severity
Word-aligned prosody is increasingly used to detect depression and posttraumatic stress disorder (PTSD), yet the stability and clinical relevance of these measurements remain unclear.AV-FlowEdit: Training-Free Prompt-Driven Audio-Visual Editing with Event-Aware Spectral Trajectory Preservation
Joint audio-video diffusion models generate synchronized content, but editing existing clips requires target changes without disrupting non-target structure or cross-modal correspondence.From 2.1M Hours of Raw Broadcast Audio to Finnish Speech SSL: Confidence Filtering and Architectural Analysis at Scale
Broadcast archives offer scale and diversity for self-supervised learning (SSL) but contain substantial noisy, non-speech, and multilingual content.Does Location Matter? A Black-Box Causal Audit Reveals Model-Dependent Audio Grounding in LALM Hallucination
Large audio-language models (LALMs) frequently exhibit object hallucination, affirming absent sounds.What Do Audio-Language Judges of Generated Music Listen For? Four Probes with Controls from Expert Ratings
Open audio-language models are used as judges of generated music: given a clip and its text prompt, the model rates the clip, and the judge is validated by how well its ratings correlate with expert…EmoGavel: Grounding-Aware Reinforcement Learning with Verifiable Rewards for Multimodal Emotion Preference Learning
Multimodal emotion preference learning uses pairwise comparisons to identify which open-ended emotion descriptions align more closely with human judgments.