AI summary

This paper introduces Connectionist Temporal Classification (CTC), a novel method using RNNs for labeling unsegmented sequences. It removes the need for pre-segmented data and post-processing. Applied to the TIMIT speech corpus, CTC outperforms baseline HMM and hybrid HMM-RNN models.

Abstract

Many real-world sequence learning tasks require the prediction of sequences of labels from noisy, unsegmented input data. In speech recognition, for example, an acoustic signal is transcribed into words or sub-word units. Recurrent neural networks (RNNs) are powerful sequence learners that would seem well suited to such tasks. However, because they require pre-segmented training data, and post-processing to transform their outputs into label sequences, their applicability has so far been limited. This paper presents a novel method for training RNNs to label unsegmented sequences directly, thereby solving both problems. An experiment on the TIMIT speech corpus demonstrates its advantages over both a baseline HMM and a hybrid HMM-RNN.