Gram-CTC: Automatic Unit Selection and Target Decomposition for Sequence Labelling

By Hairong Liu, Zhenyao Zhu, Xiangang Li, Sanjeev Satheesh. 2017

56 citations recorded by the source corpus.

Most existing sequence labelling models rely on a fixed decomposition of a target sequence into a sequence of basic units. These methods suffer from two major drawbacks: 1) the set of basic units is fixed, such as the set of words, characters or phonemes in speech recognition, and 2) the decomposition of target sequences is fixed. These drawbacks usually result in sub-optimal performance of modeling sequences. In this pa- per, we extend the popular CTC loss criterion to alleviate these limitations, and propose a new loss function called Gram-CTC. While preserving the advantages of CTC, Gram-CTC automatically learns the best set of basic units (grams), as well as the most suitable decomposition of tar- get sequences. Unlike CTC, Gram-CTC allows the model to output variable number of characters at each time step, which enables the model to capture longer term dependency and improves the computational efficiency. We demonstrate that the proposed Gram-CTC improves CTC in terms of both performance and efficiency on the large vocabulary speech recognition task at multiple scales of data, and that with Gram-CTC we can outperform the state-of-the-art on a standard speech benchmark.

Subfields: Artificial Intelligence, Signal Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Signal Processing

Citations in the corpus

Letter-Based Speech Recognition with Gated ConvNets (2017, neutral): “7 (Liu et al., 2017)† 6.”
Improved training of end-to-end attention models for speech recognition (2018, methodology): “Connectionist temporal classiﬁcation (CTC) [10] has been often used as an end-to-end model for speech recognition, often on characters/graphemes [11–16] or on sub-words [17] but also directly on words [18,19].”
English Conversational Telephone Speech Recognition by Humans and Machines (2017, neutral): “Finally, we note that the feature fusion LSTM compares favorably with other single acoustic models from the literature as mentioned in [21] (Table 4).”
Exploring neural transducers for end-to-end speech recognition (2017, neutral): “O ur s CTC [16] 9.”
Building Competitive Direct Acoustics-to-Word Models for English Conversational Speech Recognition (2017, methodology): “This includes work on character models [9, 27] and N-grams [28], RNN-Transducer [29], RNN-Aligner [30].”
Recent progresses in deep learning based acoustic models (2017, methodology): “In [69], gram-CTC was proposed to automatically learn the most suitable decomposition of target sequences.”
Improving Attention-Based End-to-End ASR Systems with Sequence-Based Loss Functions (2018, background): “Some followup work have integrated character dependencies directly into the model, such as recurrent neural network transducer (RNN-transducer) [22, 20] and recurrent neural aligner [23, 24].”
Differentiable Weighted Finite-State Transducers (2020, background): “Learning n -gram word decompositions with both differentiable (Liu et al., 2017) and non-differentiable (Chan et al., 2017) loss functions has also been explored.”
End-to-End Speech Recognition: A Survey (2023, background): “…assecondaryapproachesoutsidetheE2Eobjective,evenmore soifacousticdataisconsideredforsubwordderivation[33], [34],[35],[36]. f) Generic Modeling: Finally, E2E modeling also requires genericity of the underlying modeling: task-speciﬁc constraints are learned…”
Improved Knowledge Distillation from Bi-Directional to Uni-Directional LSTM CTC for End-to-End Speech Recognition (2018, background): “Instead, E2E ASR only requires pairs of input feature sequences and output symbol sequences [5, 6], such as phones [7], characters [8], words [9, 10], or their combinations [11, 12].”
A comprehensive analysis on attention models (2019, background): “47 CTC [Liu et al., 2017] Gram-CTC none word piece 16.”