Data processing: speech signal processing – linguistics – language – Speech signal processing – Application
Reexamination Certificate
1999-01-28
2002-05-07
Korzuch, William (Department: 2741)
Data processing: speech signal processing, linguistics, language
Speech signal processing
Application
C704S260000, C704S251000
Reexamination Certificate
active
06385586
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Technical Field
The present invention relates in general to speech communication, and in particular to a method and system for improving speech communication in foreign languages. Still more particularly, the present invention relates to a method and system for providing a language translation environment on a data processing system to improve speech communication in foreign languages.
2. Description of the Related Art
Human speech communication is often very complex, particularly when it involves communicating with someone who speaks a different language. As the world quickly evolves more and more, like a global city, with many people visiting foreign countries or conducting business with foreign partners communication in other languages becomes increasingly common. A need exists for fast and easily accessed translation environments.
Traditionally, language translation was conducted via the services of a human translator. Utilization of a human translator is rather costly and is now out-moded. Additionally, it is often difficult to find a translator on the spur of the moment which is when most translation is needed. In today's fast paced environment, individuals cannot wait around until a translator is found to communicate.
Modern technological advancement has provided us with machine translators. Machine translators are typically data processing systems configured for translation utilizing language translation software. AltaVista (™) Translation utilizing Systran technology is one of the more common web locations which provide a language translation service. AltaVista's Translation Assistant provides a free service to the Internet community. Systran, a leader in the field of machine translations, provides a machine translation of the message received from one language to another selected language. Systran Software, Inc. is a leading developer of natural language machine translation, having developed 28 language pairs. Machine translators translate only in text form, since most current language translation applications operate with ASCII text files. Therefore, all speech communication must first be converted to text before it can be translated.
Human speech voice recognition technology has been around for more than several years and is well known in the art and commercially available. Speech analysis and speech recognition algorithms, machines, and devices are becoming more and more common. Such systems have become increasingly powerful and less expensive. Those familiar with the technology are aware that various applications exist which recognize human speech and store it in various forms on a data processing system. One extension of this technology is in speech-to-text application which provides a textual representation on a data processing system of human speech.
Speech-to-text applications have evolved as one of the major goals of speech recognition technology. Many current applications utilize this technology to convert spoken language into text form which is then made accessible to a user on a data processing system.
Some common examples of the implementation of voice recognition and speech to text technology are Dragon™ (a product of COREL) and ViaVoice™ and IBM Voicetype™, both products of International Business Machines Corporation (IBM).
ViaVoice 98 Executive Edition is IBM's most powerful continuous speech software. The ViaVoice Executive 98 software package offers direct dictation into most popular Windows applications, voice navigation of your desktop and applications and the utilization of intuitive “natural language commands” for editing and formatting Microsoft Word documents.
Outputting of human speech in other than text form is also often desired. This has led to the creation of text-to-speech applications and related technological advancements. The converse of voice recognition is voice synthesizing wherein a user of a data processing system outputs a human speech signal. Human speech generation is quickly developing and is also well known in the art. In text-to speech applications, a speech synthesizer provides a computer with the ability to speak. Users and applications provide text to a speech synthesizer, which is then converted to audio.
A machine which can rapidly and automatically translate languages, particularly speech has been sought for many years. Local devices have been developed to allow instant language translations; however, attempting to put all of the capabilities of a translation environment into a local device is not feasible. Additionally, local devices tend to be huge and cumbersome and rather inefficient for someone on the go. They also have low capacity. Laptops are examples of possible local devices, however laptops are also cumbersome and tend to be rather expensive.
Most machine translation packages have been designed as an aid for professional translations, and produce outputs which have to be post edited before being presentable in their target languages. Most packages operate in a slow batch processing mode, neither of which is suitable for real time speech operation.
Many times there is a need to communicate in languages we do not speak or write, e.g. to ask directions, order food or communicate an illness or accident while traveling. A language capture device that could tie into a language translation environment with more power, capabilities and the ability to improve as better technology evolved would be much better.
It is therefore desirable to have a language capture device and system that allows for near instantaneous translation of voice inputs (speech) to enable efficient communication in a foreign language. It is further desirable to have such a device which is easily transportable and relatively inexpensive.
SUMMARY OF THE INVENTION
It is therefore one object of the present invention to provide improved speech communication.
It is another object of the present invention to provide a method and system for improved speech communication in foreign languages.
It is yet another object of the present invention to provide a method and system for providing a language translation environment on a data processing system to improve speech communication in foreign languages.
The foregoing objects are achieved as is now described. A method is disclosed for dynamically providing language translations of a human utterance from a first human language into a second human language. The method first captures the human utterance in a first human language utilizing a speech input device. The speech input device is then linked to a server created from components including a data processing system equipped with software enabled speech recognition environment and language translation environment. A desired second human language is then selected for said first human language to be translated into. Following this selection, the captured human utterance is transmitted to the server where it is converted into text utilizing the speech recognition engine of the server which instantiates the translation of the text from the first human language into the desired second human language. Finally, an output is provided of the captured human utterance in its desired second human language.
In one embodiment of the invention, a default second human language is selected based on global positioning system (GPS) data from the speech input devices equipped with GPS technology. This default language may be overridden by a user.
The above as well as additional objects, features, and advantages of the present invention will become apparent in the following detailed written description.
REFERENCES:
patent: 4882681 (1989-11-01), Brotz
patent: 5373441 (1994-12-01), Hirai et al.
patent: 5384701 (1995-01-01), Stentiford et al.
patent: 5640575 (1997-06-01), Maruyama et al.
patent: 5875422 (1999-02-01), Eslambolchi et al.
patent: 6161082 (2000-12-01), Goldberg et al.
patent: 6173250 (2001-01-01), Jong
patent: 6175819 (2001-01-01), Van Alstine
patent: 6182026 (2001-01-01), Tillmann et al.
patent: 6233544 (2001-05-01), Ak
Abebe Daniel
Bracewell & Patterson L.L.P.
International Business Machines - Corporation
Korzuch William
Leeuwen Leslie A. Van
LandOfFree
Speech recognition text-based language conversion and... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Speech recognition text-based language conversion and..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Speech recognition text-based language conversion and... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2827542