| Paper: | SP-P4.4 | ||
| Session: | Topics in Speech Understanding Systems | ||
| Time: | Tuesday, May 18, 15:30 - 17:30 | ||
| Presentation: | Poster | ||
| Topic: | Speech Processing: Spoken Language Systems and Dialog | ||
| Title: | A DISTRIBUTED FRAMEWORK FOR ENTERPRISE LEVEL SPEECH RECOGNITION SERVICES | ||
| Authors: | Iker Arizmendi; AT&T Labs - Research | ||
| Richard Rose; McGill University | |||
| Abstract: | This paper presents methods for improving the efficiency of automatic speech recognition (ASR) decoders in multi-user applications. The methods involve allocating ASR resources to service human-machine dialogs in deployments that make use of many low cost, commodity servers. It is shown that even very simple strategies for efficient allocation of ASR servers to incoming utterances has the potential to increase the capacity of a multi-user deployment. This is important because, while there has been a great deal of work applied to increasing the efficiency of individual ASR engines, there has been little effort applied to increasing overall effiency at peak loads in multi-user scenarios. Both theoretically predicted and actual performance measured on a multi-user deployment servicing a large volume of simultaneous calls demonstrated increased efficiency of nearly a factor of two over existing implementations. | ||
| Back | |||
Home -||-
Organizing Committee -||-
Technical Committee -||-
Technical Program -||-
Plenaries
Paper Submission -||-
Special Sessions -||-
ITT -||-
Paper Review -||-
Exhibits -||-
Tutorials
Information -||-
Registration -||-
Travel Insurance -||-
Housing -||-
Workshops