Mixture-of-Experts with Intermediate CTC Supervision for Accented Speech Recognition

• URL: http://arxiv.org/abs/2602.01967v1
• Date: Mon, 02 Feb 2026 11:16:34 GMT
• Title: Mixture-of-Experts with Intermediate CTC Supervision for Accented Speech Recognition
• Abstract summary: We introduce Moe-Ctc, a Mixture-of-Experts architecture with intermediate CTC supervision that jointly promotes expert specialization and generalization.<n>During training, accent-aware routing encourages experts to capture accent-specific patterns, which gradually transitions to label-free routing for inference.<n>Experiments on the Mcv-Accent benchmark demonstrate consistent gains across both seen and unseen accents in low- and high-resource conditions, achieving up to 29.3% relative WER reduction over strong FastConformer baselines.
• Score: 21.893229843453458
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Accented speech remains a persistent challenge for automatic speech recognition (ASR), as most models are trained on data dominated by a few high-resource English varieties, leading to substantial performance degradation for other accents. Accent-agnostic approaches improve robustness yet struggle with heavily accented or unseen varieties, while accent-specific methods rely on limited and often noisy labels. We introduce Moe-Ctc, a Mixture-of-Experts architecture with intermediate CTC supervision that jointly promotes expert specialization and generalization. During training, accent-aware routing encourages experts to capture accent-specific patterns, which gradually transitions to label-free routing for inference. Each expert is equipped with its own CTC head to align routing with transcription quality, and a routing-augmented loss further stabilizes optimization. Experiments on the Mcv-Accent benchmark demonstrate consistent gains across both seen and unseen accents in low- and high-resource conditions, achieving up to 29.3% relative WER reduction over strong FastConformer baselines.

Related papers

• Contrastive Regularization for Accent-Robust ASR [5.302303646066551]
  We investigate supervised contrastive learning (SupCon) as a lightweight, accent-invariant auxiliary objective for CTC fine-tuning.<n>SupCon provides an effective and model-agnostic regularization strategy for improving accent robustness.
  arXiv  Detail & Related papers  (2026-05-05T02:35:44Z)
• Pay Attention to CTC: Fast and Robust Pseudo-Labelling for Unified Speech Recognition [61.39209522608919]
  Unified Speech Recognition has emerged as a semi-supervised framework for training a single model for audio, visual, and audiovisual speech recognition.<n>We propose CTC-driven teacher forcing, where greedily decoded CTC pseudo-labels are fed into the decoder to generate attention targets.<n>Because CTC and CTC-driven attention pseudo-labels have the same length, the decoder can predict both simultaneously.
  arXiv  Detail & Related papers  (2026-02-22T19:38:21Z)
• HENT-SRT: Hierarchical Efficient Neural Transducer with Self-Distillation for Joint Speech Recognition and Translation [19.997594859651233]
  HENT-SRT is a novel framework that factorizes ASR and translation tasks to better handle reordering.<n>We improve computational efficiency by incorporating best practices from ASR transducers.<n>Our approach is evaluated on three conversational datasets Arabic, Spanish, and Mandarin.
  arXiv  Detail & Related papers  (2025-06-02T18:37:50Z)
• 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)
• Qifusion-Net: Layer-adapted Stream/Non-stream Model for End-to-End Multi-Accent Speech Recognition [1.0690007351232649]
  We propose a layer-adapted fusion (LAF) model, called Qifusion-Net, which does not require any prior knowledge about the target accent. Experiment results demonstrate that our proposed methods outperform the baseline with relative reductions of 22.1$%$ and 17.2$%$ in character error rate (CER) across multi accent test datasets.
  arXiv  Detail & Related papers  (2024-07-03T11:35:52Z)
• Accented Speech Recognition With Accent-specific Codebooks [53.288874858671576]
  Speech accents pose a significant challenge to state-of-the-art automatic speech recognition (ASR) systems. Degradation in performance across underrepresented accents is a severe deterrent to the inclusive adoption of ASR. We propose a novel accent adaptation approach for end-to-end ASR systems using cross-attention with a trainable set of codebooks.
  arXiv  Detail & Related papers  (2023-10-24T16:10:58Z)
• Multi-pass Training and Cross-information Fusion for Low-resource End-to-end Accented Speech Recognition [12.323309756880581]
  Low-resource accented speech recognition is one of the important challenges faced by current ASR technology. We propose a Conformer-based architecture, called Aformer, to leverage both the acoustic information from large non-accented and limited accented training data.
  arXiv  Detail & Related papers  (2023-06-20T06:08:09Z)
• Audio-Visual Efficient Conformer for Robust Speech Recognition [91.3755431537592]
  We propose to improve the noise of the recently proposed Efficient Conformer Connectionist Temporal Classification architecture by processing both audio and visual modalities. Our experiments show that using audio and visual modalities allows to better recognize speech in the presence of environmental noise and significantly accelerate training, reaching lower WER with 4 times less training steps.
  arXiv  Detail & Related papers  (2023-01-04T05:36:56Z)
• CTC Alignments Improve Autoregressive Translation [145.90587287444976]
  We argue that CTC does in fact make sense for translation if applied in a joint CTC/attention framework. Our proposed joint CTC/attention models outperform pure-attention baselines across six benchmark translation tasks.
  arXiv  Detail & Related papers  (2022-10-11T07:13:50Z)
• Investigating the Reordering Capability in CTC-based Non-Autoregressive End-to-End Speech Translation [62.943925893616196]
  We study the possibilities of building a non-autoregressive speech-to-text translation model using connectionist temporal classification (CTC) CTC's success on translation is counter-intuitive due to its monotonicity assumption, so we analyze its reordering capability. Our analysis shows that transformer encoders have the ability to change the word order.
  arXiv  Detail & Related papers  (2021-05-11T07:48:45Z)

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.