Dragondictate (TM)-30k: natural language speech recognition with 30, 000 words
نویسنده
چکیده
The DragonDictateTM-30K automatic transcription system is a !arge vocabulary, discrete utterance, natural language speech recognition system. First-time users are not required to provide any initial training or enrollment speech data before using the system. Speech models for 16,000 frequent words I phrases of English are built-in. Powerful adaptation techniques dynamically build and refine up to 30,000 speech models on-line during system use. With its integration of a 80,000 word dictionary and easy introduction of new terminology, DragonDictate TM offers a !arge, open vocabulary which can be used to generate free text immediately in any discipline, however specialized. Although slower than a skilled typist, interactive users of the DragonDictateTM system typically create printed text more quickly by speaking than they can generate handwritten text alone. Compatible with most popular and custom software for word-processing, databases, spreadsheets, etc., DragonDictateTM-30K presently consists of I 8-bit peripheral card and software, running near real-time on an MS-DOS 386-based (AT-bus), personal computer with 6MB memory. Professionals, business executives, and other people with limited typing skills, are now independently creating their documents and reports simply by speakin~ them. This paper describes the features of DragonDictateT from both technical and user perspectives. SYSTEM DESIGN
منابع مشابه
Dragon
Objectives: Dragon Systems is developing and building high performance, computationallyefficient interactive speech workstations, to sup'port adaptive speech recognition of large vocabulary, natural language speech input in real-time. These systems are based on the results of the multi-knowledge source (MKS) algorithm architecture and multi-processor accelerator hardware studies previously unde...
متن کاملHandling OOV words in Arabic ASR via flexible morphological constraints
We propose a novel framework to detect and recognize outof-vocabulary (OOV) words in automated speech recognition (ASR). In the proposed framework a hybrid language model combining words and sub-word units is incorporated during ASR decoding then three different OOV words recognition methods are applied to generate OOV word hypotheses. Specifically, dictionary lookup, morphological composition,...
متن کاملBulgarian National Corpus Project
The paper presents Bulgarian National Corpus project (BulNC) a large-scale, representative, online available corpus of Bulgarian. The BulNC is also a monolingual general corpus, fully morpho-syntactically (and partially semantically) annotated, and manually provided with detailed meta-data descriptions. Presently the Bulgarian National corpus consists of about 320 000 000 graphical words and in...
متن کاملSpoken Term Detection for Persian News of Islamic Republic of Iran Broadcasting
Islamic Republic of Iran Broadcasting (IRIB) as one of the biggest broadcasting organizations, produces thousands of hours of media content daily. Accordingly, the IRIBchr('39')s archive is one of the richest archives in Iran containing a huge amount of multimedia data. Monitoring this massive volume of data, and brows and retrieval of this archive is one of the key issues for this broadcasting...
متن کاملSpeech Comparison In The Rosetta Stone TM
The Rosetta Stone TM is a successful CDROM based interactive program for teaching foreign languages, that uses speech comparison to help students improve their pronunciation. The input to a speech comparison system is N+I digitised utterances. The output is a measure of the similarity of the last utterance to each of the N others. Which language is being spoken is irrelevant. This differs from ...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 1989