Applying SPHINX-II to the DARPA Wall Street Journal CSR Task

Applying SPHINX-II to the DARPA Wall Street Journal CSR Task F. Alleva, H. Hon, X. Huang, M. Hwang, R. Rosenfeld, R. Weide School of Computer Science Carnegie Mellon University Pittsburgh, Pennsylvania 15213 ABSTRACT This paper reports recent efforts to apply the speaker-independent SPHINX-H system to the DARPA Wall Street Journal continuous speech recognition task. In SPHINX-H, we incorporated ad- ditional dynamic and speaker-normalized features, replaced dis- crete models with sex-dependent semi-continuous hidden Markov models, augmented within-word triphones with between-word triphones, and extended generalized triphone models to shared- distribution models. The configuration of SPHINX-II being used for this task includes sex-dependent, semi-continuous, shared- distribution hidden Markov models and left context dependent between-word triphones. In applying our technology to this task we addressed issues that were not previously of concern owing to the (relatively) small size of the Resource Management task. 1 1. Introduction Extending a continuous speech recognition system to a larger vocabulary and more general task domain requires more than a new dictionary and language model. The primary problem in the application of the SPHINX-II [1] system to the Wall Street Journal (WSJ) CSR task was to extend the Viterbi beam-search used in the SPHINX [2] system to be able to run experiments given the constraints of available processing and memory resources. First, we developed a practical form of between-word co- articulation modelng that was both time and memory ef- ficient The use of left context dependent between-word triphones is a departure from the left and fight between- word context modeling but it allows the system to retain partial between-word co-articulation modelng despite the size and complexity of the task. Second, we significantly reduced the size of the memory required. To reduce the memory requirements of our search component it was necessary to change the Viterbi evaluation to use an in- 1This research is spomm~'ed by the Defense Advanced Research Projects Aganey, DoD, through J~pA Order 7239, and monitored by the Space and Naval Warfare Systems Cenamand under contract N00039-91-C-0158. Views and cenclusions contained in this document are those of the authors and should not be interpreted as representing official polieie~, either expxessed or implied, of the Defense Advanced Research Pr'ojeets Agency or of the United States Government. place algorithm instead of a non-in-place one. Additionally we replaced the stack data structure used to recover the word sequence from the search, with a dictionary data structure. We decoupled the proto-type HMM state tran- sition probabilities from the word specific HMM instances to avoid duplicating memory. We also found that our pointerless implementation of the HMM topology saved us both memory and time. Finally, we improved decoding efficiency substantially. One way to improve decoder ef- ficiency is to reduce the search space. SPHINX-II reduces the search space with three pruning thresholds that are ap- plied at the state, model, and word levels. In addition, evaluating a state requires an acoustic score computation and a graph update operation. Both of these operations run in constant time over one state. For discrete models, …

Content truncated to 500 words