Browse Prior Art Database

Method and System for Speech-to-Speech Translation over Mobile Phones

IP.com Disclosure Number: IPCOM000199945D
Publication Date: 2010-Sep-21
Document File: 3 page(s) / 60K

Publishing Venue

The IP.com Prior Art Database

Abstract

A method and system for speech-to-speech translation over mobile phones is disclosed. The method reduces memory and computational requirements for speech-to-speech translation over mobile phones.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 51% of the total text.

Page 1 of 3

Method and System for Speech-to-Speech Translation over Mobile Phones

Disclosed is a method and system for speech-to-speech translation over mobile phones. Current softwares for speech-to-speech translation are installed locally on a system and require a high computation capability and large memory. Thus, speech to speech translation on mobiles phones is difficult due to the requirements of high computational capabilities and large memory. Additionally, two speakers talk to one phone, thereby increasing the requirement for high computational capabilities and large memory as two directions of translations need to be performed on each phone.

The method and system disclosed herein, includes two or more mobile phones. On each mobile phone one or more of an Automatic Speech Recognition (ASR) and Text-To-Speech (TTS) translator is installed. The ASR and TTS translator are only for a native language associated with the user of the mobile phone. Translation components for translating the native language to a required language may either run on a server or the translation components corresponding to one direction of translation may be installed on the mobile phone. Thus, the user's speech in the native language of the user is recognized using the ASR and the speech is transmitted to a server to be translated to the required language. Similarly, text in the native language is first converted to speech and subsequently transmitted to the server to be translated.

With reference to Fig. 1, an English ASR and TTS are installed on a mobile phone of an English speaker. Thereafter, the speech is transmitted to an English-to-Chinese translation server. The translation components running on the English-to-Chinese translation server translate the English speech to Chinese. In other words, the English speaker's speech in the English language is recognized using the ASR and the speech is transmitted to the English-to-Chinese translation server to be translated to Chinese. Similarly, a Chinese ASR and TTS are installed on a mobile phone of a Chinese speaker. Thereafter, the Chinese speaker's speech is recognized using the Chinese ASR and TTS and subsequently transmitted to a Chinese-to-English translation server to be translated to English. As illustrated, the translation process is bi-directional and is...