Variational Connectionist Temporal Classification For Order-preserving Sequence Modeling
2023 · Zheng Nan, Ting Dang, Vidhyasaharan Sethu, et al.
Abstract
Connectionist temporal classification (CTC) is commonly adopted for sequence modeling tasks like speech recognition, where it is necessary to preserve order between the input and target sequences. However, CTC is only applied to deterministic sequence models, where the latent space is discontinuous and sparse, which in turn makes them less capable of handling data variability when compared to variational models. In this paper, we integrate CTC with a variational model and derive loss functions that can be used to train more generalizable sequence models that preserve order. Specifically, we derive two versions of the novel variational CTC based on two reasonable assumptions, the first being that the variational latent variables at each time step are conditionally independent; and the second being that these latent variables are Markovian. We show that both loss functions allow direct optimization of the variational lower bound for the model log-likelihood, and present computationally t
Authors
(none)
Tags
Stats
Related papers
- A Study Of All-convolutional Encoders For Connectionist Temporal Classification (2017)5.84
- Training LDCRF Model On Unsegmented Sequences Using Connectionist Temporal Classification (2016)2.26
- Comparison Of Decoding Strategies For CTC Acoustic Models (2017)10.48
- Order-preserving Abstractive Summarization For Spoken Content Based On Connectionist Temporal Classification (2017)3.58
- CR-CTC: Consistency Regularization On CTC For Improved Speech Recognition (2024)6.30
- Investigating The Reordering Capability In Ctc-based Non-autoregressive End-to-end Speech Translation (2021)0.00
- Adamer-ctc: Connectionist Temporal Classification With Adaptive Maximum Entropy Regularization For Automatic Speech Recognition (2024)5.84
- Align With Purpose: Optimize Desired Properties In CTC Models With A General Plug-and-play Framework (2023)0.00