TheanoLM - An Extensible Toolkit for Neural Network Language Modeling

Abstract

We present a new tool for training neural network language models (NNLMs), scoring sentences, and generating text. The tool has been written using Python library Theano, which allows researcher to easily extend it and tune any aspect of the training process. Regardless of the flexibility, Theano is able to generate extremely fast native code that can utilize a GPU or multiple CPU cores in order to parallelize the heavy numerical computations. The tool has been evaluated in difficult Finnish and English conversational speech recognition tasks, and significant improvement was obtained over our best back-off n-gram models. The results that we obtained in the Finnish task were compared to those from existing RNNLM and RWTHLM toolkits, and found to be as good or better, while training times were an order of magnitude shorter.

References in corpus (4)

• Adam: A Method for Stochastic Optimization
• ADADELTA: An Adaptive Learning Rate Method
• Theano: new features and speed improvements
• Techniques for effective vocabulary selection

Cited by in corpus (6)

• A Survey on Neural Network Language Models
• Automatic Speech Recognition with Very Large Conversational Finnish and Estonian Vocabularies
• A Character-Word Compositional Neural Language Model for Finnish
• A Multifaceted Evaluation of Neural versus Phrase-Based Machine Translation for 9 Language Directions
• Code-switching Language Modeling With Bilingual Word Embeddings: A Case Study for Egyptian Arabic-English
• Effects of Language Relatedness for Cross-lingual Transfer Learning in Character-Based Language Models