Portability Issues for Speech Recognition Technologies

Portability Issues for Speech Recognition Technologies � Lori Lamel, Fabrice Lefevre, Jean-Luc Gauvain and Gilles Adda Spoken Language Processing Group, CNRS-LIMSI, 91403 Orsay, France flamel,lefevre,gauvain,gaddag@limsi.fr ABSTRACT Although there has been regular improvement in speech recog- nition technology over the past decade, speech recognition is far from being a solved problem. Most recognition systems are tuned to a particular task and porting the system to a new task (or lan- guage) still requires substantial investment of time and money, as well as expertise. Todays state-of-the-art systems rely on the avail- ability of large amounts of manually transcribed data for acous- tic model training and large normalized text corpora for language model training. Obtaining such data is both time-consuming and expensive, requiring trained human annotators with substantial a- mounts of supervision. In this paper we address issues in speech recognizer portabil- ity and activities aimed at developing generic core speech recogni- tion technology, in order to reduce the manual effort required for system development. Three main axes are pursued: assessing the genericity of wide domain models by evaluating performance under several tasks; investigating techniques for lightly supervised acous- tic model training; and exploring transparent methods for adapting generic models to a specific task so as to achieve a higher degree of genericity. 1. INTRODUCTION The last decade has seen impressive advances in the capability and performance of speech recognizers. Todays state-of-the-art systems are able to transcribe unrestricted continuous speech from broadcast data with acceptable performance. The advances arise from the increased accuracy and complexity of the models, which are closely related to the availability of large spoken and text cor- pora for training, and the wide availability of faster and cheaper computational means which have enabled the development and im- plementation of better training and decoding algorithms. Despite the extent of progress over the recent years, recognition accuracy is still extremely sensitive to the environmental conditions and speak- ing style: channel quality, speaker characteristics, and background �This work was partially financed by the European Commission under the IST-1999 Human Language Technologies project 11876 Coretex. . noise have an important impact on the acoustic component of the speech recognizer, whereas the speaking style and the discourse domain have a large impact on the linguistic component. In the context of the EC IST-1999 11876 project CORETEX we are investigating methods for fast system development, as well as development of systems with high genericity and adaptability. By fast system development we refer to: language support, i.e., the capability of porting technology to different languages at a reason- able cost; and task portability, i.e. the capability to easily adapt a technology to a new task by exploiting limited amounts of domain- specific knowledge. Genericity and adaptability refer to the capac- ity of the technology to work properly on a wide range of tasks and to dynamically keep models up to date using contemporary data. The more robust the initial generic system is, the less there is a need for adaptation. Concerning the acoustic modeling component, genericity …

Content truncated to 500 words