Recurrent neural network language model adaptation for multi-genre broadcast speech recognition and alignment

Recurrent neural network language models (RNNLMs) generally outperform n-gram language models when used in automatic speech recognition. Adapting RNNLMs to new domains is an open problem and current approaches can be categorised as either feature-based and model-based. In feature-based adaptation, the input to the RNNLM is augmented with auxiliary features whilst model-based adaptation includes model fine-tuning and the introduction of adaptation layer(s) in the network. In this paper, the properties of both types of adaptation are investigated on multi-genre broadcast speech recognition. Existing techniques for both types of adaptation are reviewed and the proposed techniques for model-based adaptation, namely the linear hidden network (LHN) adaptation layer and the K-component adaptive RNNLM, are investigated. Moreover, new features derived from the acoustic domain are investigated for RNNLM adaptation. The contributions of this paper include two hybrid adaptation techniques: the fine-tuning of feature-based RNNLMs and a feature-based adaptation layer. Moreover, the semi-supervised adaptation of RNNLMs using genre information is also proposed. The ASR systems were trained using 700h of multi-genre broadcast speech. The gains obtained when using the RNNLM adaptation techniques proposed in this work are consistent when using RNNLMs trained on an in-domain set of 10M words and on a combination of in-domain and out-of-domain sets of 660M words, with approx. 10% perplexity and 2% relative word error rate improvements on a 28.3h. test set. The best RNNLM adaptation techniques for ASR are also evaluated on a lightly supervised alignment of subtitles task for the same data, where the use of RNNLM adaptation leads to an absolute increase in the F-measure of 0.5%.