arXiv AI
Jun 30
Explainable AI in Speaker Recognition -- Making Latent Representations Understandable
arXiv:2604. 23354v3 Announce Type: replace-cross Abstract: Neural networks can be trained to learn task-relevant representations from data.
Exploring Second-Order Pattern Recognition in Speaker Recognition
The paper introduces the concept of second‑order patterns—latent structures that underlie a speaker recognition network’s classification of utterances. Using hierarchical clustering, the authors identify these patterns and interpret them with the HCCM method. They then propose a new task, second‑order pattern recognition, and present the HCNA method, which improves performance by matching unseen utterances to the extrapolation space of identified clusters.
Related stories
arXiv AI
Sep 15
Interpreting hierarchical organisation of speaker embeddings
arXiv:2609.15203v1 Announce Type: cross Abstract: Speaker recognition neural networks learn latent representations (i.e. speaker embeddings) from input utterances to recognise speaker identities. How...
arXiv Computation and Language
Sep 22
Enhancing speech representation learning with cross-modal knowledge transfer with HGNN under low resource settings: the case study of Yemba
arXiv:2609.23194v1 Announce Type: new Abstract: Acoustic representation learning is crucial for speech processing, yet low-resource languages (LRLs) face severe data scarcity, limiting the effectiven...
arXiv Machine Learning
2d ago
Training-Free Affinity Fusion of Neural and Embedding-Based Speaker Diarization
arXiv:2609.39162v1 Announce Type: cross Abstract: Speaker diarization systems based on speaker embeddings and neural diarization exploit complementary forms of speaker information, but their intermed...
arXiv AI
Jul 21
Unsupervised Multimodal Clustering for Semantics Discovery in Multimodal Utterances
arXiv:2405. 12775v2 Announce Type: replace-cross Abstract: Discovering the semantics of multimodal utterances is essential for understanding human language and enhancing human-machine interactions.
arXiv Computation and Language
3d ago
VOSSA: Voiceprint Optimization for Streaming Speech Architectures
arXiv:2609.38887v1 Announce Type: cross Abstract: Real-time voice conversion (VC) systems commonly rely on pretrained speaker embeddings from automatic speaker verification (ASV) models. While effect...
arXiv AI
Jun 15
Listening with Attention: Entropy-Guided Explainability for Transformer-Based Audio Models
arXiv:2606. 14647v1 Announce Type: cross Abstract: Transformer-based automatic speech recognition (ASR) models such as Whisper are highly accurate, but their predictions remain difficult to interpret.
arXiv AI
Jul 10
PS4: Proxy-Supervised Joint Training for Real Target Speaker Extraction
arXiv:2607. 08111v1 Announce Type: cross Abstract: Training target speaker extraction (TSE) models for real conversational mixtures remains challenging because large-scale training corpora and clean target speech for supervision are unavailable.
arXiv AI
Jun 15
HyperPotter: Spell the Charm of High-Order Interactions in Audio Deepfake Detection
arXiv:2602. 05670v2 Announce Type: replace-cross Abstract: Advances in AIGC technologies have enabled the synthesis of highly realistic audio deepfakes capable of deceiving human auditory perception.
arXiv AI
Jul 24
AG-REPA: Causal Layer Selection for Representation Alignment in Audio Flow Matching
arXiv:2603. 01006v3 Announce Type: replace-cross Abstract: REPresentation Alignment (REPA) improves the training of generative flow models by aligning intermediate hidden states with pretrained teacher features, but its effectiveness in token-conditioned audio Flow Matching critically depends on the choice of supervised layers, which is typically made heuristically based on the depth.
arXiv AI
Jul 28
Extracting Algorithms in Pre-trained LLMs: A Case on Hidden Markov Models
arXiv:2607. 22646v1 Announce Type: new Abstract: Large language models (LLMs) display a striking ability to predict next observations from Hidden Markov Models (HMMs) via in-context learning (ICL), but the algorithm underlying this capability remains undetermined: prior work has proposed several candidates without consensus, and none has been grounded in the model's internal activations.
arXiv Machine Learning
Aug 7
MoDAl: Self-Supervised Neural Modality Discovery via Decorrelation for Speech Neuroprosthesis
arXiv:2605. 00025v3 Announce Type: replace-cross Abstract: Speech neuroprosthesis systems decode intended speech from neural activity in the absence of audible output, offering a path to restoring communication for individuals with speech-impairing conditions.