arXiv Machine Learning

Revisiting Accent Classification with DistilHuBERT: Speaker-Disjoint Evaluation and Cross-Corpus Generalization

The paper evaluates the robustness of DistilHuBERT for accent classification using speaker‑disjoint splits and cross‑corpus tests with L2‑ARCTIC and other English accent datasets. It finds a large performance gap between in‑corpus and external evaluations, indicating that file‑level tests overestimate accuracy. The study shows that balanced external‑data training performs best, while pseudo‑labeling with confidence filtering is insufficient under domain mismatch.

arXiv Machine Learning

Aug 28

Benchmarking_Fast_Domain_Adaptation_for_Unsupervised_Speech_Units

The paper introduces ABX-Accent, a benchmark built on the AESRC dataset that evaluates how well representation learning models adapt to 10 different English accents with less than 10 hours of unlabeled data per accent. It adapts the Zero Resources Challenge ABX metrics for each accent and demonstrates a baseline using adaptive domain normalization to fine‑tune a Contrastive Predictive Coding model, achieving a 23.6% relative improvement on across‑speaker ABX scores compared to non‑adapted models. The dataset and evaluation metrics will be released publicly after the paper is accepted.

By Robin San Roman, Manel Khentout, Tu Anh Nguyen, Paul Michel, Yossi Adi, Emmanuel Dupoux

benchmarks

Hugging Face Trending Papers

Aug 27

Benchmarking_Fast_Domain_Adaptation_for_Unsupervised_Speech_Units

The paper introduces ABX-Accent, a benchmark for evaluating how well unsupervised speech representation learning models adapt to new accents. It uses the AESRC dataset with 10 English accents, each providing less than 10 hours of unlabeled data, and adapts the Zero Resources Challenge ABX metrics to these accents. A baseline model employing adaptive domain normalization to fine‑tune a Contrastive Predictive Coding model shows a 23.6% relative improvement on across‑speaker ABX scores compared to non‑adapted models.

benchmarks

arXiv Machine Learning

Oct 7

AccentCL: Robust Accent Classification with Incremental Expansion

arXiv:2610.07426v1 Announce Type: cross Abstract: Accent classifiers are typically trained with a fixed label inventory and cannot accommodate new accent categories as new data becomes available. Mor...

By Mu-Ruei Tseng, Waris Quamer, Ghady Nasrallah, Ricardo Gutierrez-Osuna

multimodal safety

arXiv Machine Learning

Jun 19

PASQA: Pitch-Accent-Focused Speech Quality Assessment Model Trained on Synthetic Speech with Accent Errors

arXiv:2606. 20137v1 Announce Type: cross Abstract: Existing mean opinion score (MOS) prediction models typically predict utterance-level naturalness MOS and can be insensitive to localized pitch-accent errors.

By Masaya Kawamura, Yuma Shirahata, Kentaro Mitsui, Reo Shimizu

multimodal

arXiv AI

Sep 15

Modeling, Scaling, and Decoding: Optimizing Controllable Speech Generation with Nonverbal Vocalizations

arXiv:2609.14231v1 Announce Type: cross Abstract: Controllable synthesis of nonverbal vocalizations (NVVs) is es- sential for natural and expressive speech, but remains challeng- ing due to their aco...

By Ziyu Zhang, Yun Chen, Taihui Wang, Hanzhao Li, Qicong Xie, Rilin Chen, Zhixian Zhao, Lei Xie

fine-tuning

arXiv Machine Learning

Jun 16

Contrastive Regularization for Accent-Robust ASR

arXiv:2605. 03297v2 Announce Type: replace-cross Abstract: ASR systems based on self-supervised acoustic pretraining and CTC fine-tuning achieve strong performance on native speech but remain sensitive to accent variability.

By Van-Phat Thai, Aradhya Dhruv, Duc-Thinh Pham, Sameer Alam

fine-tuning multimodal benchmarks

arXiv Machine Learning

Jun 26

Neural Speaker Diarization via Multilingual Training: Evaluation on Low-Resource Nepali-Hindi Speech

arXiv:2606. 26144v1 Announce Type: cross Abstract: Speaker diarization, the task of determining "who spoke when" in a multi-speaker recording, is a critical component in applications such as meeting transcription, accessibility tools, and multilingual information retrieval.

By Samip Neupane, Sandesh Pokhrel, Sandesh Pyakurel, Basanta Joshi

safety

arXiv Computation and Language

Oct 8

Mitigating Accent-Language Confusion in Self-Supervised Speech Representations for Language Identification

arXiv:2610.09486v1 Announce Type: cross Abstract: Spoken language identification (LID) aims to recognize the target language regardless of accent. In practice, however, LID models fine-tuned from sel...

By Minu Kim, Jihwan Lee, David R. Mortensen, Shrikanth Narayanan

fine-tuning safety

arXiv Machine Learning

Jun 30

BabyHuBERT: Multilingual Self-Supervised Learning for Segmenting Speakers in Child-Centered Long-Form Recordings

arXiv:2509. 15001v3 Announce Type: replace-cross Abstract: Child-centered daylong recordings are essential for studying early language development, but existing speech models trained on clean adult data perform poorly due to acoustic and linguistic differences.

By Th\'eo Charlot, Tarek Kunze, Maxime Poli, Alejandrina Cristia, Emmanuel Dupoux, Marvin Lavechin

arXiv AI

Jul 28

Indic DiarBench: A Multilingual Joint Diarization and ASR Benchmark for Indian Languages

arXiv:2607. 23808v1 Announce Type: cross Abstract: In this work, we introduce Indic DiarBench, a speaker diarization and ASR benchmark dataset spanning all 22 scheduled languages of India.

By Deovrat Mehendale, Aditya Mehndiratta, Dhruv Rathi, Kaushal Bhogale, Mitesh M. Khapra

llms multimodal benchmarks

arXiv Machine Learning

Jun 9

From A to B to A: Palindromic Zero-Shot Voice Conversion with Non-Parallel Data

arXiv:2606. 08843v1 Announce Type: cross Abstract: We present a voice conversion (VC) framework that utilizes K-Nearest Neighbors (KNN) retrieval over WavLM representations to align non-parallel source and target speech, constructing synthetic training pairs for supervised learning.

By Moshe Mandel, Shlomo E. Chazan

safety

Hugging Face Trending Papers

Aug 12

Confucius4-TTS: Transcript-Free Cross-Lingual Zero-Shot TTS with a Learnable Speaker Encoder

Recent advances in zero-shot text-to-speech (TTS) have substantially improved speech quality and voice cloning fidelity. However, many zero-shot TTS systems still depend on audio prompt transcripts at inference time.

llms multimodal benchmarks

Related stories