REVIEW 5 cited by
CR-CTC: Consistency regularization on CTC for improved speech recognition
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
Connectionist Temporal Classification (CTC) is a widely used method for automatic speech recognition (ASR), renowned for its simplicity and computational efficiency. However, it often falls short in recognition performance. In this work, we propose the Consistency-Regularized CTC (CR-CTC), which enforces consistency between two CTC distributions obtained from different augmented views of the input speech mel-spectrogram. We provide in-depth insights into its essential behaviors from three perspectives: 1) it conducts self-distillation between random pairs of sub-models that process different augmented views; 2) it learns contextual representation through masked prediction for positions within time-masked regions, especially when we increase the amount of time masking; 3) it suppresses the extremely peaky CTC distributions, thereby reducing overfitting and improving the generalization ability. Extensive experiments on LibriSpeech, Aishell-1, and GigaSpeech datasets demonstrate the effectiveness of our CR-CTC. It significantly improves the CTC performance, achieving state-of-the-art results comparable to those attained by transducer or systems combining CTC and attention-based encoder-decoder (CTC/AED). We release our code at https://github.com/k2-fsa/icefall.
Forward citations
Cited by 5 Pith papers
-
Hybrid Decoding: Rapid Pass and Selective Detailed Correction for Sequence Models
A hybrid decoding scheme with a fast TDT draft decoder and selective transformer patches matches baseline word error rate while cutting decoder latency roughly threefold.
-
DESign: Dynamic Context-Aware Convolution and Efficient Subnet Regularization for Continuous Sign Language Recognition
A sign language recognition model using context-aware dynamic convolutions and subnetwork CTC regularization reports new state-of-the-art word error rates on PHOENIX14, PHOENIX14-T, and CSL-Daily.
-
HENT-SRT: Hierarchical Efficient Neural Transducer with Self-Distillation for Joint Speech Recognition and Translation
A hierarchical transducer with self-distillation and a tuned blank penalty improves joint speech recognition and translation, matching offline attention models on conversational data.
-
TVTA: Trajectory-Aware Viseme-Guided Temporal Aggregation for Event-Based Lip Reading
Trajectory-aware local temporal modeling before spatial aggregation plus CTC viseme supervision and EMA consistency lifts DVS-Lip word accuracy to 77.49%.
-
NIM4-ASR: Towards Efficient, Robust, and Customizable Real-Time LLM-Based ASR
NIM4-ASR delivers SOTA ASR performance on public benchmarks using a 2.3B-parameter LLM with multi-stage training, real-time streaming, and million-scale hotword customization via RAG.
Discussion (0). Continue with ORCID to comment.