A Comparison Of Lattice-free Discriminative Training Criteria For Purely Sequence-trained Neural Network Acoustic Models

Abstract

In this work, three lattice-free (LF) discriminative training criteria for purely sequence-trained neural network acoustic models are compared on LVCSR tasks, namely maximum mutual information (MMI), boosted maximum mutual information (bMMI) and state-level minimum Bayes risk (sMBR). We demonstrate that, analogous to LF-MMI, a neural network acoustic model can also be trained from scratch using LF-bMMI or LF-sMBR criteria respectively without the need of cross-entropy pre-training. Furthermore, experimental results on Switchboard-300hrs and Switchboard+Fisher-2100hrs datasets show that models trained with LF-bMMI consistently outperform those trained with plain LF-MMI and achieve a relative word error rate (WER) reduction of 5% over competitive temporal convolution projected LSTM (TDNN-LSTMP) LF-MMI baselines.

Stats

• citations3
• S2 citations—
• github stars0
• HF likes0
• heat score4.52
• arxiv keyweng2018a

Related papers

• Comparison Of Lattice-free And Lattice-based Sequence Discriminative Training Criteria For LVCSR (2019)5.84
• Consistent Training And Decoding For End-to-end Speech Recognition Using Lattice-free MMI (2021)8.35
• Bayesian Learning Of LF-MMI Trained Time Delay Neural Networks For Speech Recognition (2020)8.82
• Lattice-based Lightly-supervised Acoustic Model Training (2019)0.00
• On The Relation Between Internal Language Model And Sequence Discriminative Training For Neural Transducers (2023)0.00
• On Lattice-free Boosted MMI Training Of HMM And Ctc-based Full-context ASR Models (2021)7.81
• Unsupervised Model-based Speaker Adaptation Of End-to-end Lattice-free MMI Model For Speech Recognition (2022)2.26
• A Novel Pyramidal-fsmn Architecture With Lattice-free MMI For Speech Recognition (2018)0.00