A tree-based statistical language model for natural language speech recognition
L.R. Bahl, Peter F. Brown, P.V. de Souza et al. · IEEE Transactions on Acoustics Speech and Signal Processing · 1989
The problem of predicting the next word a speaker will say, given the words already spoken; is discussed. Specifically, the problem is to estimate the probability that a given word will be the next word uttered. Algorithms are presented for automatically constructing a binary decision tree designed to estimate these probabilities. At each node of the tree there is a yes/no question relating to the words already spoken, and at each leaf there is a probability distribution over the allowable vocabulary. Ideally, these nodal questions can take the form of arbitrarily complex Boolean expressions, but computationally cheaper alternatives are also discussed. Some results obtained on a 5000-word vocabulary with a tree designed to predict the next word spoken from the preceding 20 words are included. The tree is compared to an equivalent trigram model and shown to be superior.>
Cited by 255 works. doi:10.1109/29.32278
Authors
References
- Optimization by Simulated Annealing
- A Maximum Likelihood Approach to Continuous Speech Recognition
- Classification and regression trees
- Prediction and Entropy of Printed English
- Continuous speech recognition by statistical methods
- An Algorithm for Constructing Optimal Binary Decision Trees
- Application of information theory to the construction of efficient decision trees
- The Development of an Experimental Discrete Dictation Recognizer
- The development of an experimental discrete dictation recognizer
- An information theoretic approach to the automatic determination of phonemic baseforms