mlsanityData to 7 Oct 2026

Sequence-to-Sequence Speech Recognition with Time-Depth Separable Convolutions

arXiv 1904.02619 ↗First posted 4 Apr 2019cs.CLcited by 27

Not on the map. Why?

Abstract (arXiv)

We propose a fully convolutional sequence-to-sequence encoder architecture with a simple and efficient decoder. Our model improves WER on LibriSpeech while being an order of magnitude more efficient than a strong RNN baseline. Key to our approach is a time-depth separable convolution block which dramatically reduces the number of parameters in the model while keeping the receptive field large. We also give a stable and efficient beam search inference procedure which allows us to effectively integrate a language model. Coupled with a convolutional language model, our time-depth separable convolution architecture improves by more than 22% relative WER over the best previously reported sequence-to-sequence results on the noisy LibriSpeech test set.

Key facts. Cited by 27 papers in the map (data through 7 Oct 2026). First posted on arXiv 4 Apr 2019.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 7; 2022: 4; 2023: 3; 2024: 4; 2025: 5; 2026: 4.

27 papers.

Sort:

  1. 2021VoxPopuli: A Large-Scale Multilingual Speech Corpus for Representation Learning, Semi-Supervised Learning and Interpretationcited by 455
  2. 2023End-to-End Speech Recognition: A Surveycited by 127
  3. 2024emg2pose: A Large and Diverse Benchmark for Surface Electromyographic Hand Pose Estimationcited by 20
  4. 2024emg2qwerty: A Large Dataset with Baselines for Touch Typing using Surface Electromyographycited by 17
  5. 2021Acoustic Data-Driven Subword Modeling for End-to-End Speech Recognitioncited by 11
  6. 2021Capturing Event Argument Interaction via A Bi-Directional Entity-Level Recurrent Decodercited by 11
  7. 2021Unsupervised Automatic Speech Recognition: A Reviewcited by 5
  8. 2022Supervised Attention in Sequence-to-Sequence Models for Speech Recognitioncited by 3
  9. 2023RASR2: The RWTH ASR Toolkit for Generic Sequence-to-sequence Speech Recognitioncited by 3
  10. 2024LLM-Driven Multimodal Opinion Expression Identificationcited by 3
  11. 2025emg2speech: Synthesizing speech from electromyography using self-supervised speech modelscited by 3
  12. 2021Deconvolution-and-convolution Networkscited by 2
15 more
  1. 2021Compressing 1D Time-Channel Separable Convolutions using Sparse Random Ternary Matricescited by 2
  2. 2023SpeeChain: A Speech Toolkit for Large-Scale Machine Speech Chaincited by 2
  3. 2025Let SSMs be ConvNets: State-space Modeling with Optimal Tensor Contractionscited by 2
  4. 2025Scaling and Distilling Transformer Models for sEMGcited by 2
  5. 2022Partitioned Gradient Matching-based Data Subset Selection for Compute-Efficient Robust ASR Trainingcited by 1
  6. 2024Self-consistent context aware conformer transducer for speech recognitioncited by 1
  7. 2025SplashNet: Split-and-Share Encoders for Accurate and Efficient Typing with Surface Electromyographycited by 1
  8. 2026EgoEMG: A Multimodal Egocentric Dataset with Bilateral EMG and Vision for Hand Pose Estimationcited by 1
  9. 2021Voice Quality and Pitch Features in Transformer-Based Speech Recognitioncited by 0
  10. 2022Mixture-of-Rookies: Saving DNN Computations by Predicting ReLU Outputscited by 0
  11. 2022A Comparison of Transformer, Convolutional, and Recurrent Neural Networks on Phoneme Recognitioncited by 0
  12. 2025Typing Reinvented: Towards Hands-Free Input via sEMGcited by 0
  13. 2026From Muscle to Text with MyoText: sEMG to Text via Finger Classification and Transformer-Based Decodingcited by 0
  14. 2026Myovox: Reading Speech from the Muscles of the Facecited by 0
  15. 2026EFormer: Temporally Aligned Local Correction for Continuous sEMG-Based Hand Pose Trackingcited by 0