End-to-end speech recognition using lattice-free MMI

We present our work on end-to-end training of acoustic models using the lattice-free maximum mutual information (LF-MMI) objective function in the context of hidden Markov models. By end-to-end training, we mean flat-start training of a single DNN in one stage without using any previously trained models, forced alignments, or building state-tying decision trees... We use full biphones to enable context-dependent modeling without trees, and show that our end-to-end LF-MMI approach can achieve comparable results to regular LF-MMI on well-known large vocabulary tasks. We also compare with other end-to-end methods such as CTC in character-based and lexicon-free settings and show 5 to 25 percent relative reduction in word error rates on different large vocabulary tasks while using significantly smaller models. read more

PDF Abstract

Datasets


  Add Datasets introduced or used in this paper

Results from the Paper


Task Dataset Model Metric Name Metric Value Global Rank Benchmark
Speech Recognition Switchboard (300hr) End-to-end LF-MMI Word Error Rate (WER) 9.3 # 1
Speech Recognition WSJ eval92 End-to-end LF-MMI Word Error Rate (WER) 3.0 # 3

Methods


No methods listed for this paper. Add relevant methods here