mlsanityData to 7 Oct 2026

CTC Variations Through New WFST Topologies

arXiv 2110.03098 ↗First posted 6 Oct 2021eess.AS cs.CL cs.LGcited by 17

Not on the map. Why?

Abstract (arXiv)

This paper presents novel Weighted Finite-State Transducer (WFST) topologies to implement Connectionist Temporal Classification (CTC)-like algorithms for automatic speech recognition. Three new CTC variants are proposed: (1) the "compact-CTC", in which direct transitions between units are replaced with <epsilon> back-off transitions; (2) the "minimal-CTC", that only adds <blank> self-loops when used in WFST-composition; and (3) the "selfless-CTC" variants, which disallows self-loop for non-blank units. Compact-CTC allows for 1.5 times smaller WFST decoding graphs and reduces memory consumption by two times when training CTC models with the LF-MMI objective without hurting the recognition accuracy. Minimal-CTC reduces graph size and memory consumption by two and four times for the cost of a small accuracy drop. Using selfless-CTC can improve the accuracy for wide context window models.

Key facts. Cited by 17 papers in the map (data through 7 Oct 2026). Builds on 13 of its 23 references that are in the map. First posted on arXiv 6 Oct 2021.

Builds on

13 of 23 references.

References · 13

  1. 2020Conformer: Convolution-augmented Transformer for Speech Recognitioncited by 1,838
  2. 2019NeMo: a toolkit for building AI applications using Neural Modulescited by 230
  3. 2015EESEN: End-to-End Speech Recognition using Deep RNN Models and WFST-based Decodingcited by 73
  4. 2021Citrinet: Closing the Gap between Non-Autoregressive and Autoregressive End-to-End Models for Automatic Speech Recognitioncited by 33
  5. 2021Why does CTC result in peaky behavior?cited by 32
  6. 2019From Senones to Chenones: Tied Context-Dependent Graphemes for Hybrid Speech Recognitioncited by 20
  7. 2020Semi-Supervised Speech Recognition via Graph-based Temporal Classificationcited by 19
  8. 2020PyChain: A Fully Parallelized PyTorch Implementation of LF-MMI for End-to-End ASRcited by 16
  9. 2020Benchmarking LF-MMI, CTC and RNN-T Criteria for Streaming ASRcited by 11
  10. 2020Faster, Simpler and More Accurate Hybrid ASR Systems Using Wordpiecescited by 9
  11. 2019GPU-Accelerated Viterbi Exact Lattice Decoder for Batched Online and Offline Speech Recognitioncited by 4
  12. 2021Advancing CTC-CRF Based End-to-End Speech Recognition with Wordpieces and Conformerscited by 4
1 more
  1. 2019A Novel Topology for End-to-end Temporal Classification and Segmentation with Recurrent Neural Networkcited by 1

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 3; 2023: 10; 2024: 0; 2025: 3; 2026: 1.

17 papers.

Sort:

By area:

  1. 2023End-to-End Speech Recognition: A Surveycited by 127
  2. 2023DiffVoice: Text-to-Speech with Latent Diffusioncited by 13
  3. 2022Bayes risk CTC: Controllable CTC alignment in Sequence-to-Sequence taskscited by 12
  4. 2023Text-only domain adaptation for end-to-end ASR using integrated text-to-mel-spectrogram generatorcited by 11
  5. 2022Star Temporal Classification: Sequence Classification with Partially Labeled Datacited by 10
  6. 2023Bypass Temporal Classification: Weakly Supervised Automatic Speech Recognition with Imperfect Transcriptscited by 10
  7. 2023Blank-regularized CTC for Frame Skipping in Neural Transducercited by 8
  8. 2023Learning from Flawed Data: Weakly Supervised Automatic Speech Recognitioncited by 8
  9. 2023Powerful and Extensible WFST Framework for RNN-Transducer Lossescited by 7
  10. 2023Weakly-supervised forced alignment of disfluent speech using phoneme-level modelingcited by 7
  11. 2023Alternative Pseudo-Labeling for Semi-Supervised Automatic Speech Recognitioncited by 6
  12. 2023GPU-Accelerated WFST Beam Search Decoder for CTC-based Speech Recognitioncited by 4
5 more
  1. 2022Blank Collapse: Compressing CTC emission for the faster decodingcited by 3
  2. 2025Enhancing GOP in CTC-Based Mispronunciation Detection with Phonological Knowledgecited by 2
  3. 2025RNN-Transducer-based Losses for Speech Recognition on Noisy Targetscited by 1
  4. 2025Phonetically-Augmented Discriminative Rescoring for Voice Search Error Correctioncited by 0
  5. 2026Modeling Overlapped Speech with Shufflescited by 0