Benchmarking_Fast_Domain_Adaptation_for_Unsupervised_Speech_Units

• URL: http://arxiv.org/abs/2608.26992v1
• Date: Thu, 27 Aug 2026 11:40:18 GMT
• Title: Benchmarking_Fast_Domain_Adaptation_for_Unsupervised_Speech_Units
• Abstract summary: We introduce ABX- Accent, a benchmark based on the AESRC dataset that features 10 different accents of English.<n>It includes a small ( 10 hours) unlabelled training set in each of the accents and adaptations of the Zero Resources Challenge ABX evaluation metrics to each of the accents.<n>When applied to the new benchmark, the proposed method yields a relative improvement of 23.6% on across-speaker ABX scores on average.
• Score: 39.74450851447742
• License: http://creativecommons.org/licenses/by-sa/4.0/
• Abstract: Representation learning has attracted great atten- tion and managed to reach good performances as a pretraining method for downstream tasks or as a first step towards unsu- pervised speech modeling. Yet, little is known about how such methods deal with out-of-domain speech and how could they be adapted in a few shot to new domains. This is important especially for accented speech where one observes a long tail of accents that diverge from the standard ones. We introduce ABX- Accent, a benchmark based on the AESRC dataset that features 10 different accents of English. It includes a small (< 10 hours) unlabelled training set in each of the accents and adaptations of the Zero Resources Challenge ABX evaluation metrics to each of the accents. We illustrate this benchmark with a baseline model that uses adaptive domain normalization to fine tune a pretrained Contrastive Predictive Coding model on the accents. This method is first developed on LibriSpeech using a male/female split. When applied to the new benchmark, the proposed method yields a relative improvement of 23.6% on across-speaker ABX scores on average compared to non adapted models. The data and metrics will be open sourced upon paper acceptance

Related papers

• SpidR-Adapt: A Universal Speech Representation Model for Few-Shot Adaptation [40.55805997909858]
  We introduce SpidR-Adapt for rapid adaptation to new languages using minimal unlabeled data.<n>We construct a multi-task adaptive pre-training protocol which formulates the adaptation process as a bi-level optimization framework.<n> Empirically, SpidR-Adapt achieves rapid gains in phonemic discriminability and spoken language modeling.
  arXiv  Detail & Related papers  (2025-12-24T14:33:16Z)
• CLAIR-A: Leveraging Large Language Models to Judge Audio Captions [73.51087998971418]
  evaluating machine-generated audio captions is a complex task that requires considering diverse factors.<n>We propose CLAIR-A, a simple and flexible method that leverages the zero-shot capabilities of large language models.<n>In our evaluations, CLAIR-A better predicts human judgements of quality compared to traditional metrics.
  arXiv  Detail & Related papers  (2024-09-19T17:59:52Z)
• Improving Self-supervised Pre-training using Accent-Specific Codebooks [48.409296549372414]
  accent-aware adaptation technique for self-supervised learning. On the Mozilla Common Voice dataset, our proposed approach outperforms all other accent-adaptation approaches.
  arXiv  Detail & Related papers  (2024-07-04T08:33:52Z)
• Self-Taught Recognizer: Toward Unsupervised Adaptation for Speech Foundation Models [84.8919069953397]
  Self-TAught Recognizer (STAR) is an unsupervised adaptation framework for speech recognition systems. We show that STAR achieves an average of 13.5% relative reduction in word error rate across 14 target domains. STAR exhibits high data efficiency that only requires less than one-hour unlabeled data.
  arXiv  Detail & Related papers  (2024-05-23T04:27:11Z)
• Improving Speech Recognition for African American English With Audio Classification [17.785482810741367]
  We propose a new way to improve the robustness of a US English short-form speech recognizer using a small amount of out-of-domain data. Fine-tuning on this data results in a 38.5% relative word error rate disparity reduction between AAE and MAE without reducing MAE quality.
  arXiv  Detail & Related papers  (2023-09-16T19:57:45Z)
• Continual Learning for On-Device Speech Recognition using Disentangled Conformers [54.32320258055716]
  We introduce a continual learning benchmark for speaker-specific domain adaptation derived from LibriVox audiobooks. We propose a novel compute-efficient continual learning algorithm called DisentangledCL. Our experiments show that the DisConformer models significantly outperform baselines on general ASR.
  arXiv  Detail & Related papers  (2022-12-02T18:58:51Z)
• Cross-domain Voice Activity Detection with Self-Supervised Representations [9.02236667251654]
  Voice Activity Detection (VAD) aims at detecting speech segments on an audio signal. Current state-of-the-art methods focus on training a neural network exploiting features directly contained in the acoustics. We show that representations based on Self-Supervised Learning (SSL) can adapt well to different domains.
  arXiv  Detail & Related papers  (2022-09-22T14:53:44Z)
• Low-resource Accent Classification in Geographically-proximate Settings: A Forensic and Sociophonetics Perspective [8.002498051045228]
  Accented speech recognition and accent classification are relatively under-explored research areas in speech technology. Recent deep learning-based methods and Transformer-based pretrained models have achieved superb performances in both areas. In this paper, we explored three main accent modelling methods combined with two different classifiers based on 105 speaker recordings retrieved from five urban varieties in Northern England.
  arXiv  Detail & Related papers  (2022-06-26T01:25:17Z)
• Sequence-level self-learning with multiple hypotheses [53.04725240411895]
  We develop new self-learning techniques with an attention-based sequence-to-sequence (seq2seq) model for automatic speech recognition (ASR) In contrast to conventional unsupervised learning approaches, we adopt the emphmulti-task learning (MTL) framework. Our experiment results show that our method can reduce the WER on the British speech data from 14.55% to 10.36% compared to the baseline model trained with the US English data only.
  arXiv  Detail & Related papers  (2021-12-10T20:47:58Z)
• Evaluating the reliability of acoustic speech embeddings [10.5754802112615]
  Speech embeddings are fixed-size acoustic representations of variable-length speech sequences. Here, we systematically compare two popular metrics, ABX discrimination and Mean Average Precision (MAP), on 5 languages across 17 embedding methods. We find that overall, ABX and MAP correlate with one another and with frequency estimation.
  arXiv  Detail & Related papers  (2020-07-27T13:24:09Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.