Data processing: speech signal processing – linguistics – language – Speech signal processing – Application
Reexamination Certificate
2001-01-18
2004-11-23
Abebe, Daniel (Department: 2655)
Data processing: speech signal processing, linguistics, language
Speech signal processing
Application
C704S257000
Reexamination Certificate
active
06823312
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates to a personalized system for providing a service for improving understandability of received speech in accordance with user specific needs. The said system is online and used by a plurality of users, addressing the user's inability to understand speech.
BACKGROUND OF THE INVENTION
The existing solutions are all in the form of an equipment or device that can be used only by one person. The problem with such individual-use devices is that it is not feasible and practical for each such individual device to stay continuously upgraded with the latest advancements in technology or to dynamically customize with the changes in the user's acoustic profile, usage environment and conversation context. There are multiple reasons for this. It is also not always possible to customize an off-the-shelf equipment for an individual's disability and needs. Also the latest technological advancements and algorithms are likely to be expensive for incorporation in an individual device, thereby limiting its quality of service. A device like this is usually required to be used for a long period of time, in some cases for the lifetime of the individual. It is not easy for a device to adjust and customize dynamically to the changes in an individuals disability over a period of time, without requiring a repurchase. It is also not possible to make use of the specific conversation context or environment to achieve better results. E.g. the user could be using the device in a plurality of business contexts, in social setting or at home during the day. It is not easy to customize an individuals device at such fine granularity level.
Some systems have been proposed that address other aspects of speech understanding. For example U.S. Pat. No. 6,036,496 describes an apparatus and method for screening an individual's ability to process acoustic events. The invention provides sequences (or trials) of acoustically processed target and distracter phoneme to a subject for identification. The acoustic processing includes amplitude emphasis of selected frequency envelopes, stretching (in the time domain) of selected portions of phoneme, and phase adjustment of selection portions of phoneme relative to a base frequency. After a number of trials, the invention develops a profile for an individual that indicates whether the individual's ability to process acoustic events is within a normal range, and if not, what processing can provide the individual with optimal hearing. The invention provides a method to determine an individual's acoustic profile. This is better than the typical hearing tests, which determine whether an individual can hear particular frequencies, at particular amplitudes. The invention also mentions that the individual's profile can then be used by a listening or processing device to particularly emphasize, stretch, or otherwise manipulate an audio stream to provide the individual with an optimal chance of distinguishing between similar acoustic events.
Another U.S. Pat. No. 6,071,123 proposes a method and a system that provides means to enable individuals with speech, language and reading based communication disabilities, due to a temporal processing problem, to improve their temporal processing abilities as well as their communication abilities. The method and system include provisions to elongate portions of phoneme that have brief and/or rapidly changing acoustic spectra, such as occur in the stop consonants b and d in the phonemes /ba/ and /da/, as well as reduce the duration of the steady state portion of the syllable. In addition, some emphasis is added to the rapidly changing segments of these phonemes. Additionally, the disclosure includes method for and computer software to modify fluent speech to make the modified speech better recognizable by communicatively impaired individuals. The proposed apparatus is a device or an equipment to be used by an individual.
U.S. Pat. No. 6,109,107 provides an improved method and apparatus for the identification and treatment of language perception problems in specific language impaired (SLI) individuals. The invention provides a method and apparatus for screening individuals for SLI and training individuals who suffer from SLI to re-mediate the effects of the impairment by using the spectral content of interfering sound stimuli and the temporal ordering or direction of the interference between the stimuli. This emphasis in this invention is on screening and training individuals and not providing a device or a service to address the disability.
U.S. Pat. No. 5,839,109 also describes a speech recognition apparatus that includes a sound pickup, a standard feature storage device, a comparing device, a display pattern storing device, and a display. The apparatus can display non-speech sounds either as a message or as an image, and is especially useful for hearing-impaired individuals. For example, if a fire engine siren is detected, the display can show a picture of a fire engine, or can display the message “siren is sounding”.
All of the above solutions are limited to addressing hearing disabilities and are not directed at improving the understandability of speech which is an issue that could occur even with individuals without hearing disabilities. For example aspects relating to spoken accent or as an extreme case, a different language are not addressed by any of the above solutions.
In addition, even for cases where physical disability is involved, none of the above solutions addresses those situations where extreme disabilities occur—for Example, complete loss of hearing or complete loss of hearing coupled with blindness.
The existing solutions are also non-adaptive as they do not automatically adjust to dynamically varying individual requirements-eg. Ambient noise levels, change in hearing patterns etc., nor are they capable of automatically adapting to different user profiles, as a result it is not feasible for multiple users to use the same system.
DETAILED DESCRIPTION
The object of this invention is to obviate the above drawbacks and to provide personalized improved understandability of speech based on an individual's needs.
The second object of this invention is to display the speech in text or as graphics on a display panel on the phone device instead of being an audio heard through the phone speaker.
Another object of this invention is to provide data processing functionality as a third party service to a plurality of users, over a network, such as an Intranet, an Extranet or an Internet.
Yet another object of this invention is to provide a self learning system using artificial intelligence and expert system techniques.
Another object of this invention is to provide a speech-enabled WAP (Wireless Application Protocol) system for hearing or speech.
To achieve the said objective this invention provides a personalized system for providing a service for improving understandability of received speech in accordance with user specific needs characterized in that it includes:
input interface means for capturing received speech signals connected to a speech recognition or speech signal analysis means for identifying the contents of the received speech connected to one input of a data processing means for performing improvement in understandability,
a user profile storage means connected to another input of said data processing means for providing user specific improvement data, and
an output generation means connected to the output of said data processing means to produce personalized output based on an individual's needs.
The said personalized system is online.
The said speech recognition means is any known speech recognition means.
The said data processing means is a computing system.
The said data processing means is a server system in a client server environment.
The said data processing means is a self-learning system using artificial intelligence or expert system techniques, which improves its performance based on feedback from the users over a period of t
Dubey Pradeep Kumar
Mittal Parul A.
Abebe Daniel
Coca, Esq. T. Rao
International Business Machines - Corporation
McGinn & Gibb PLLC
LandOfFree
Personalized system for providing improved understandability... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Personalized system for providing improved understandability..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Personalized system for providing improved understandability... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3357146