KR20020024045A - An unseen interpretation & practice speaking foreign language make use of cellular phone - Google Patents

Abstract

PURPOSE: A system for supporting foreign language conversation study and interpretation using a cellular phone is provided to allow a user to be able to study conversation and interpret through a voice recognition system without inputting character data. CONSTITUTION: The voice of a speaker is inputted and converted into a digital signal, to be spooled in a host. A standardization recognition module corresponding to personal data is executed on the basis of a personal interface. The speaker's database and an arbitrary speaker's standardization data are divided from each other. A composite voice recognition process for converting phoneme into an English word or other language and a continuous vocabulary synthesizing process for synthesizing continuous vocabularies are continuously performed to collect similar phoneme of candidate words. By doing so, at least one data is extracted, and the data and candidate vocabularies are output to a mobile terminal.

Description

An unseen interpretation & practice speaking foreign language make use of cellular phone}

The present invention is a system for a telephone information service that provides a real-time translation and conversation learning through a complex speech recognition process and a continuous vocabulary synthesis recognition process in a speech recognition system using a mobile terminal.

In general, the mobile terminal 2 has a communication module, a voice input means, a voice output means, and a call state and image data display means for transmitting and receiving wireless RF waves, undergoing modulation and demodulation, and transmitting and receiving wireless LAN waves. (6) converting any data input by the speaker through the A / D adapter (4) into personalized data (6) and changing the phonology into English words or other languages by the language processing method (8) And recognizing and synthesizing the continuous vocabulary through Viterbi search and repeating the synthesized data to the end (10) and applying the final recognized vocabulary through the output of the algorithm detected and extracted through the candidate recognition word. The method includes transmitting the above data to the mobile terminal.

According to the present invention, as part of a service for providing contents supported by a mobile terminal, the R & D following a product innovation for a new technology reduces spending on parts added in a product or appearance of a new model by the added parts. It is to maximize the utilization of existing products and to increase the utility of the voice recognition system currently used to solve the problems of improving the service quality of the product and securing the customer. Reduce the 4PS of other new products, and use the translation or translation necessary for learning English or other languages using mobile communication devices and communicating with foreigners in everyday life anytime, anywhere without the support of an individual, and the additional burden of purchasing a product I'm trying to solve the problem of. Complementing memory chips and system modules in existing communication devices or commercializing only newly released models prevents the reduction of production costs and the mass build of personal databases.

1 is a perspective view showing a conventional cellular phone

2 is a structural diagram of a system using the present invention

<Explanation of symbols for main parts of drawing>

2. Cellphone 4. A / D, D / A adapter

6. Speech Recognition System 8. Continuous Speech Recognition Process

10. Continuous Vocabulary Synthesis Process 12. Data Output Module

14. Standardized Database 16. Complex Speech Recognition Process

According to the present invention, as shown in Fig. 2, the speaker's voice is input and the analog is converted to digital through the adapter spooled to the host: a step of minimizing the variation through the normalization process and extracting the end point detection feature through a low pass filter. Distinguishing the speaker from the random speaker in the decoding step of implementing a standardized recognition module corresponding to the individual's data based on the connected individual's interface: the step of separately executing the speaker's database and the arbitrary speaker into the standardized data : Speech Recognition System applies standardized data to process multiple vocabularies by using a speech processing method that converts phonology into English words or other languages by using language processing method and discrete HMM and Viterbi algorithms. Continuous Vocabulary Synthesis Process through Synthesis Recognition According to the recognition result to the continuous phase after going through the process of set of the candidate words similar phoneme deriving at least one or more of the data and outputs it to the mobile terminal to a digital coder including candidate words.

The R & D following the innovation of the product for new technology does not require the expenditure of parts added in the products of the mobile communication devices which are currently commercially available, or the appearance of a new model by the added parts. In order to solve problems related to improving the service quality of the product and securing the customer by increasing the quality of the product, an interpreter or translation necessary for learning English or other languages and communicating with foreigners can be used in daily life. It is trying to solve the problem of burden. In this regard, the company will be able to seek to enter the global market as a new concept while realizing customer retention as a comparative advantage in the field of marketing and maximizing its utility as a specialized service.

Claims (1)

It is to implement a system that enables real-time conversational learning and interpretation for random talkers through voice recognition system without directly inputting text data using cellular phone.