CTC Variations Through New WFST Topologies

Abstract

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.

Accepted to Interspeech 2022, 5 pages, 2 figures, 7 tables

References in corpus (4)

• Citrinet: Closing the Gap between Non-Autoregressive and Autoregressive End-to-End Models for Automatic Speech Recognition
• Why does CTC result in peaky behavior?
• Advancing CTC-CRF Based End-to-End Speech Recognition with Wordpieces and Conformers
• A Novel Topology for End-to-end Temporal Classification and Segmentation with Recurrent Neural Network

Cited by in corpus (2)

• Text-only domain adaptation for end-to-end ASR using integrated text-to-mel-spectrogram generator
• Powerful and Extensible WFST Framework for RNN-Transducer Losses