Large-vocabulary speaker-independent continuous speech recognition: the sphinx system
Large-vocabulary speaker-independent continuous speech recognition: the sphinx system
A stack decoder for continous speech recognition
HLT '89 Proceedings of the workshop on Speech and Natural Language
The N-Best algorithm: an efficient procedure for finding top N sentence hypotheses
HLT '89 Proceedings of the workshop on Speech and Natural Language
A CSR-NL interface specification version 1.5
HLT '89 Proceedings of the workshop on Speech and Natural Language
Problem-Solving Methods in Artificial Intelligence
Problem-Solving Methods in Artificial Intelligence
Dragon systems resource management benchmark results—February 1991
HLT '91 Proceedings of the workshop on Speech and Natural Language
Modelling context dependency in acoustic-phonetic and lexical representations
HLT '91 Proceedings of the workshop on Speech and Natural Language
Integration of diverse recognition methodologies through reevaluation of N-best sentence hypotheses
HLT '91 Proceedings of the workshop on Speech and Natural Language
HLT '90 Proceedings of the workshop on Speech and Natural Language
An overview of the SPHINX-II speech recognition system
HLT '93 Proceedings of the workshop on Human Language Technology
Search algorithms for software-only real-time recognition with very large vocabularies
HLT '93 Proceedings of the workshop on Human Language Technology
The kernelHMM: learning kernel combinations in structured output domains
Proceedings of the 29th DAGM conference on Pattern recognition
Text entry for mobile devices using ad-hoc abbreviation
Proceedings of the International Conference on Advanced Visual Interfaces
ACLShort '10 Proceedings of the ACL 2010 Conference Short Papers
Unified stochastic engine (USE) for speech recognition
ICASSP'93 Proceedings of the 1993 IEEE international conference on Acoustics, speech, and signal processing: speech processing - Volume II
Generating phrasal and sentential paraphrases: A survey of data-driven methods
Computational Linguistics
| Hi-index | 0.00 |
In this paper a new, tree-trellis based fast search for finding the N best sentence hypotheses in continuous speech recognition is proposed. The search consists of two parts: a forward, time-synchronous, trellis search and a backward, time asynchronous, tree search. In the first module the well known Viterbi algorithm is used for finding the best hypothesis and for preparing a map of all partial paths scores time synchronously. In the second module a tree search is used to grow partial paths backward and time asynchronously. Each partial path in the backward tree search is rank ordered in a stack by the corresponding full path score, which is computed by adding the partial path score with the best possible score of the remaining path obtained from the trellis path map. In each path growing cycle, the current best partial path, which is at the top of the stack, is extended by one arc (word). The new tree-trellis search is different from the traditional time synchronous Viterbi search in its ability for finding not just the best but the N-best paths of different word content. The new search is also different from the A algorithm, or the stack algorithm, in its capability for providing an exact, full path score estimate of any given partial (i.e., incomplete) path before its completion. When compared with the best candidate Viterbi search, the search complexities for finding the N-best strings are rather low, i.e., only a fraction more computation is needed.