Audio ML Papers

Last 7 Days (February 04 - February 11, 2026)

Subcategories: All (25) | Speech Synthesis (6) | Music Synthesis (1) | Ambient Synthesis (3) | Quality Assessment (0) | Enhancement (2) | Asr (2) | Other (11)
← Previous Week | Current Week

๐Ÿ† Top Papers This Week

#1 TOP PAPER (Score: 91)
Xuenan Xu, Yiming Ren, Liwei Liu ... ยท arXiv
Recent advances in speech synthesis and editing have made speech spoofing increasingly challenging. However, most existing methods treat spoofing as binary classification, overlooking that diverse spoofing techniques manipulate multiple, coupled speech attributes and their semant...
#2 TOP PAPER (Score: 84)
Georg Heigold, Ehsan Variani, Tom Bagby ... ยท arXiv
Audio is a critical component of multimodal perception, and any truly intelligent system must demonstrate a wide range of auditory capabilities. These capabilities include transcription, classification, retrieval, reasoning, segmentation, clustering, reranking, and reconstruction...
#3 TOP PAPER (Score: 84)
Videet Mehta, Liming Wang, Hilde Kuehne ... ยท arXiv
Large audio-language models (LALMs) exhibit strong zero-shot capabilities in multiple downstream tasks, such as audio question answering (AQA) and abstract reasoning; however, these models still lag behind specialized models for certain discriminative tasks (e.g., audio classific...
Monday, February 09, 2026
Chengzhong Wang, Andong Li, Dingding Yao ... ยท arXiv
While deep learning has advanced speech enhancement (SE), effective phase modeling remains challenging, as conventional networks typically operate within a flat Euclidean feature space, which is not easy to model the underlying circular topology of the phase. To address this, we ...
Kohei Saijo, Yoshiaki Bando ยท IEEE Transactions on Audio, Speech, and Language Processing (TASLP)
Time-frequency domain dual-path models have demonstrated strong performance and are widely used in source separation. Because their computational cost grows with the number of frequency bins, these models often use the band-split (BS) module in high-sampling-rate tasks such as mu...
Haoshen Wang, Xueli Zhong, Bingbing Lin ... ยท arXiv
Dysarthric speech exhibits high variability and limited labeled data, posing major challenges for both automatic speech recognition (ASR) and assistive speech technologies. Existing approaches rely on synthetic data augmentation or speech reconstruction, yet often entangle speake...
Jiatao Chen, Xing Tang, Xiaoyue Duan ... ยท arXiv
While existing Singing Voice Synthesis systems achieve high-fidelity solo performances, they are constrained by global timbre control, failing to address dynamic multi-singer arrangement and vocal texture within a single song. To address this, we propose Tutti, a unified framewor...
Sunday, February 08, 2026
Shaad Sufi ยท arXiv
Current audio formats present a fundamental trade-off between file size and functionality: lossless formats like FLAC preserve quality but lack adaptability, while lossy formats reduce size at the cost of fidelity and offer no stem-level access.We introduce the Stem-Native Codec ...
Jiale Qian, Hao Meng, Tian Zheng ... ยท arXiv
While recent years have witnessed rapid progress in speech synthesis, open-source singing voice synthesis (SVS) systems still face significant barriers to industrial deployment, particularly in terms of robustness and zero-shot generalization. In this report, we introduce SoulX-S...
Friday, February 06, 2026
Videet Mehta, Liming Wang, Hilde Kuehne ... ยท arXiv
Large audio-language models (LALMs) exhibit strong zero-shot capabilities in multiple downstream tasks, such as audio question answering (AQA) and abstract reasoning; however, these models still lag behind specialized models for certain discriminative tasks (e.g., audio classific...
Georg Heigold, Ehsan Variani, Tom Bagby ... ยท arXiv
Audio is a critical component of multimodal perception, and any truly intelligent system must demonstrate a wide range of auditory capabilities. These capabilities include transcription, classification, retrieval, reasoning, segmentation, clustering, reranking, and reconstruction...
Yuancheng Wang, Zhenyu Tang, Yun Wang ... ยท arXiv
Speech tokenizers are foundational to speech language models, yet existing approaches face two major challenges: (1) balancing trade-offs between encoding semantics for understanding and acoustics for reconstruction, and (2) achieving low bit rates and low token rates. We propose...
Ziyu Luo, Lin Chen, Qiang Qu ... ยท arXiv
Spatial audio is crucial for creating compelling immersive 360-degree video experiences. However, generating realistic spatial audio, such as first-order ambisonics (FOA), from 360-degree videos in complex acoustic scenes remains challenging. Existing methods often overlook the d...
Hugo Seutรฉ, Pranai Vasudev, Etienne Richan ... ยท Temporary pre-print, will be updated. In review at a conference
Realistic sound propagation is essential for immersion in a virtual scene, yet physically accurate wave-based simulations remain computationally prohibitive for real-time applications. Wave coding methods address this limitation by precomputing and compressing impulse responses o...
Thursday, February 05, 2026
Chunyat Wu, Jiajun Deng, Zhengxi Liu ... ยท ICASSP 2026
Although diffusion-based, non-autoregressive text-to-speech (TTS) systems have demonstrated impressive zero-shot synthesis capabilities, their efficacy is still hindered by two key challenges: the difficulty of text-speech alignment modeling and the high computational overhead of...
Kaiyuan Zhang, Mohan Shi, Eray Eren ... ยท arXiv
Neural audio codecs are widely used for audio compression and can be integrated into token-based language models. Traditional codecs preserve acoustic details well but lack semantic information. Recent hybrid codecs attempt to incorporate semantic information through distillation...
Qing Wen, Haohao Li, Zhongjie Ba ... ยท arXiv
Advances in AIGC technologies have enabled the synthesis of highly realistic audio deepfakes capable of deceiving human auditory perception. Although numerous audio deepfake detection (ADD) methods have been developed, most rely on local temporal/spectral features or pairwise rel...
Haoqin Sun, Chenyang Lyu, Shiwan Zhao ... ยท arXiv
Despite the growing success of Large Speech Language Models (LSLMs) in processing short-term acoustic signals, their extension to long-form audio understanding is severely bottlenecked. This limitation stems from the limited context length and the exorbitant memory footprints req...
Wednesday, February 04, 2026
Xuenan Xu, Yiming Ren, Liwei Liu ... ยท arXiv
Recent advances in speech synthesis and editing have made speech spoofing increasingly challenging. However, most existing methods treat spoofing as binary classification, overlooking that diverse spoofing techniques manipulate multiple, coupled speech attributes and their semant...
Haina Zhu, Yao Xiao, Xiquan Li ... ยท arXiv
We study the fine-grained text-to-audio (T2A) generation task. While recent models can synthesize high-quality audio from text descriptions, they often lack precise control over attributes such as loudness, pitch, and sound events. Unlike prior approaches that retrain models for ...
Georgii Aparin, Tasnima Sadekova, Alexey Rukhovich ... ยท arXiv
Sparse Autoencoders (SAEs) are powerful tools for interpreting neural representations, yet their use in audio remains underexplored. We train SAEs across all encoder layers of Whisper and HuBERT, provide an extensive evaluation of their stability, interpretability, and show their...
Georgii Aparin, Tasnima Sadekova, Alexey Rukhovich ... ยท EACL 2026, main track
Sparse Autoencoders (SAEs) are powerful tools for interpreting neural representations, yet their use in audio remains underexplored. We train SAEs across all encoder layers of Whisper and HuBERT, provide an extensive evaluation of their stability, interpretability, and show their...
Tuan Dat Phuong, Duc-Tuan Truong, Long-Vu Hoang ... ยท ICASSP 2026
Transformer-based models have shown strong performance in speech deepfake detection, largely due to the effectiveness of the multi-head self-attention (MHSA) mechanism. MHSA provides frame-level attention scores, which are particularly valuable because deepfake artifacts often oc...
Amir Ivry, Shinji Watanabe ยท arXiv
Spoken dialogues with and between voice agents are becoming increasingly common, yet assessing them for their socially harmful content such as violence, harassment, and hate remains text-centric and fails to account for audio-specific cues and transcription errors. We present LAL...
Dongchao Yang, Yuanyuan Wang, Dading Chong ... ยท arXiv
We study two foundational problems in audio language models: (1) how to design an audio tokenizer that can serve as an intermediate representation for both understanding and generation; and (2) how to build an audio foundation model that generalizes in few-shot and zero-shot sett...
Dongchao Yang, Yuanyuan Wang, Dading Chong ... ยท arXiv
We study two foundational problems in audio language models: (1) how to design an audio tokenizer that can serve as an intermediate representation for both understanding and generation; and (2) how to build an audio foundation model that generalizes in few-shot and zero-shot sett...
Chien-Chun Wang, Hung-Shin Lee, Hsin-Min Wang ... ยท IEEE Transactions on Audio, Speech and Language Processing
Pre-trained models for automatic speech recognition (ASR) and speech enhancement (SE) have exhibited remarkable capabilities under matched noise and channel conditions. However, these models often suffer from severe performance degradation when confronted with domain shifts, part...
Vikentii Pankov, Artem Gribul, Oktai Tatanov ... ยท ICASSP 2026
We present PFluxTTS, a hybrid text-to-speech system addressing three gaps in flow-matching TTS: the stability-naturalness trade-off, weak cross-lingual voice cloning, and limited audio quality from low-rate mel features. Our contributions are: (1) a dual-decoder design combining ...