منابع مشابه
Recognition of spelled names over the telephone and rejection of data out of the spelling lexicon
This paper deals with the recognition of spelled names over the telephone. It introduces an efficient way of handling the spelling grammar, that is the lexicon of the allowed spelled names. The proposed approach is based on a forward-backward algorithm. The constraints on the sequences of letters are derived from the lexicon and are used by the A* algorithm in the backward pass. This forward-ba...
متن کاملEngineering out the noise
The US Navy, through an Office of Naval Research (ONR) lead effort on Noise Induced Hearing Loss (NIHL), is investigating methods and techniques to mitigate hearing loss for the crews and warfighters. Hearing protection is a viable and increasingly popular method of reducing hearing exposure for many ship crew members; however, it has limitations on comfort and low frequency effectiveness. Furt...
متن کاملRe-engineering knock-out processes
The core of many workflow processes in banks, insurance companies, governmental departments, and administrations of multinationals is formed by a set of tasks that are used to classify cases into two groups: accepted and rejected. Each of these Ž . tasks has two possible outcomes: OK or NOK i.e., Not OK . If for a specific case all tasks result in OK, the case is accepted, otherwise it is rejec...
متن کاملLanguage models for a spelled letter recognizer
In some speech recognition applications, it is reasonable to constrain the search space of a speech recognizer to a large but nite set of sentences. We demonstrate the problem on a spelling task, where the recognition of continuously spelled last names is constrained to 110,000 entries (= 43,000 unique names) of a telephone book. Several techniques to address this problem are compared: recognit...
متن کاملRecognition of spelled names over the telephone
Recognition of spelled names over the telephone line is essential for applications such as telephone directory assistance, or automatic mail ordering. We present recognition results on the spelling section of the OGI Spelled and Spoken Word Telephone Corpus, using a Multi-State Time Delay Neural Network (MS-TDNN). Many applications allow for strong language modeling constraints. In our experime...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Mechanical Engineering
سال: 2011
ISSN: 0025-6501,1943-5649
DOI: 10.1115/1.2011-nov-4