Enhancing commercial grammar-based applications using robust approaches to speech understanding
نویسنده
چکیده
This paper presents a series of measurements of the accuracy of speech understanding when grammar-based or robust approaches are used. The robust approaches considered here are based on statistical language models (SLMs) with the interpretation being carried out by phrasespotting or robust parsing methods. We propose a simple process to leverage existing grammars and logged utterances to upgrade grammar-based applications to become more robust to out-of-coverage inputs. All experiments herein are run on data collected from deployed directed dialog applications and show that SLMbased techniques outperform grammarbased ones without requiring any change in the application logic.
منابع مشابه
A robust speech understanding system using conceptual relational grammar
We describe a robust speech understanding system based on our newly developed approach to spoken language processing. We show that a robust NLU system can be rapidly developed using a relatively simple speech recognizer to provide sufficient information for database retrieval by spoken language. Our experimental system consists of three components: a speech recognizer based on HMM, a natural la...
متن کاملA methodology for comparing grammar-based and robust approaches to speech understanding
We present a series of experiments designed to compare grammar-based and robust approaches to speech understanding, performed in the context of an Open Source medical speech translation system. We used two versions of the system, one grammar-based and one robust, trained off the same training data, and evaluated them on test data collected using both versions of the system. The experiments were...
متن کاملComparing grammar-based and robust approaches to speech understanding: a case study
Previous work has demonstrated the success of statistical language models when enough training data is available [1], but despite that, grammar-based systems are proving the preferred choice in successful commercial systems such as HeyAnita [2], BeVocal [3] and Tellme [4], largely due to the difficulty involved in obtaining a corpus of training data. Here we trained an SLM on data obtained usin...
متن کاملOnline adaptation of language models in spoken dialogue systems
The robust estimation of language models for new applications of spoken dialogue systems often suffers from a shortcoming of training material. An alternative to training a language model is to improve an initial language model using material obtained while running the new system, thus adapting it to the new task. In this paper we investigate different methods for onlineadaptation of language m...
متن کاملOnline Adaptation for Language Models in Spoken Dialogue Systems
The robust estimation of language models for new applications of spoken dialogue systems often suffers from a shortcoming of training material. An alternative to training a language model is to improve an initial language model using material obtained while running the new system, thus adapting it to the new task. In this paper we investigate different methods for onlineadaptation of language m...
متن کامل