Lattice Rescoring for Speech Recognition using Large Scale Distributed Language Models
نویسندگان
چکیده
In this paper, we suggest a lattice rescoring architecture that has features of a Trie DB based language model (LM) server and a naïve parameter estimation (NPE) to integrate distributed language models. The Trie DB LM server supports an efficient computation of LM score to rerank the n-best sentences extracted from the lattice. In the case of NPE, it has a role of an integration of heterogeneous LM resources. Our approach distributes LM computations not only to distribute LM resources. This is simple and easy to implement and maintain the distributed lattice rescoring architecture. The experimental results show that the performance of the lattice rescoring has improved with the NPE algorithm that can find the optimal weights of the LM interpolation. In addition, we show that it is available to integrate n-gram LM and DIMI LM.
منابع مشابه
Fuzzy class rescoring: a part-of-speech language model
Current speech recognition systems usually use word-based trigram language models. More elaborate models are applied to word lattices or N best lists in a rescoring pass following the acoustic decoding process. In this paper we consider techniques for dealing with class-based language models in the lattice rescoring framework of our JANUS large vocabulary speech recognizer. We demonstrate how t...
متن کاملOn-the-fly lattice rescoring for real-time automatic speech recognition
This paper presents a method for rescoring the speech recognition lattices on-the-fly to increase the word accuracy while preserving low latency of a real-time speech recognition system. In large vocabulary speech recognition systems, pruned and/or lower order n-gram language models are often used in the first-pass of the speech decoder due to the computational complexity. The output word latti...
متن کاملLarge-scale random forest language models for speech recognition
The random forest language model (RFLM) has shown encouraging results in several automatic speech recognition (ASR) tasks but has been hindered by practical limitations, notably the space-complexity of RFLM estimation from large amounts of data. This paper addresses large-scale training and testing of the RFLM via an efficient disk-swapping strategy that exploits the recursive structure of a bi...
متن کاملLarge-scale discriminative language model reranking for voice-search
We present a distributed framework for largescale discriminative language models that can be integrated within a large vocabulary continuous speech recognition (LVCSR) system using lattice rescoring. We intentionally use a weakened acoustic model in a baseline LVCSR system to generate candidate hypotheses for voice-search data; this allows us to utilize large amounts of unsupervised data to tra...
متن کاملLarge Scale Language Modeling in Automatic Speech Recognition
Large language models have been proven quite beneficial for a variety of automatic speech recognition tasks in Google. We summarize results on Voice Search and a few YouTube speech transcription tasks to highlight the impact that one can expect from increasing both the amount of training data, and the size of the language model estimated from such data. Depending on the task, availability and a...
متن کامل