Journal
Publisher
ISCA-INT SPEECH COMMUNICATION ASSOC
DOI: 10.21437/Interspeech.2018-1413
Keywords
ASR; recurrent neural network language model (RNNLM); neural language model adaptation; fast marginal adaptation (FMA); cache model; deep neural network (DNN); lattice rescoring
Categories
Funding
- DARPA LORELEI [HR0011-15-2-0024]
- NSF [CRI-1513128]
- IARPA MATERIAL award [FA8650-17-C-9115]
Ask authors/readers for more resources
We propose two adaptation models for recurrent neural network language models (RNNLMs) to capture topic effects and long-distance triggers for conversational automatic speech recognition (ASR). We use a fast marginal adaptation (FMA) framework to adapt a RNNLM. Our first model is effectively a cache model the word frequencies are estimated by counting words in a conversation (with utterance-level hold-one-out) from 1st pass decoded word lattices, and then is interpolated with a background unigram distribution. In the second model, we train a deep neural network (DNN) on conversational transcriptions to predict word frequencies given word frequencies from 1st pass decoded word lattices. The second model can in principle model trigger and topic effects but is harder to train. Experiments on three conversational corpora show modest WER and perplexity reductions with both adaptation models.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available