Baiju D. Mandalia

Speaker verification for authorizing updates to user subscription service received by internet service provider (ISP) using an intelligent peripheral (IP) in an advanced intelligent network (AIN)

A method for controlling subscription services delivered to a user by an Internet Service Provider (ISP) coupled to an Automated Intelligent Network (AIN) telephone system with at least one central office switching system. An intelligent peripheral subsystem is connected to the central office switching system, via a call connection channel. The intelligent peripheral subsystem providing at least one auxiliary call processing capability via the call connection channel and provides a telephony speaker authentication method for selectively authorizing updates to user subscription services types provided by the ISP, wherein each of the subscription services types include one or more service options. The method for controlling subscription services including: a) prompting for a voice response; b) authenticating a users voice response pattern by performing a speaker-dependent cepstrum matching algorithm with previously stored samples of the users voice response, and if the users voice response pattern is positively authenticated with previously stored samples, then performing the steps of: c) presenting a menu of user subscription types, each with available AIN service options; and d) responding to a user selection of at least one user subscription types, each with service option, by activating a service option selected by the user, wherein the service option is governed by said ISP by regulating one or more Uniform Resource Locator (URL) addresses received by a user subscriber device from said ISP over a telecommunication line.

Method and apparatus for assisting speech and hearing impaired subscribers using the telephone and central office

A system and method for Assisting Speech And Hearing-impaired Subscribers Using The Telephone And Central Office is disclosed. The present invention in some preferred embodiments permits remote customization of Central Office resources on behalf of a specific subscriber, and also allows the subscriber to test the customization before deployment. The present invention helps advance the art of telephony voice assistance by adding mechanisms to solve problems present in standalone environments. The method on the system comprises the steps of: receiving a communication session request from a first party using a first user unit; and determining if the first party is speech-impaired, and if the first party is speech-impaired, then performing the sub-steps of: prompting the first party for text input; receiving the text input from the first party; converting the text input, using a Text-to-Speech resource, to an audio output; and sending the audio output to a second party.

Performance evaluation of the communications protocol processor

The performance evaluation of the protocol processing elements of a multimedia communications processor, is described. Specialized frame and header processing units were simulated using a high-level hardware model to produce typical path lengths for known protocols. A performance evaluation based on extrapolated data was performed to evaluate variable traffic and application conditions. The results indicate that such an architecture can allow handling of data for 5-10-Mb/s links without modification of existing protocols. The variation would depend on the number of layers supported.<<ETX>>

Flexible hardware architecture for multi-media communications processing

A microprocessor with features designed specifically for digital signal processing and communications protocol processing, for applications in data transmission relative to multiple media access, is described. In comparison with current microprocessors, this processor has efficient and cost-effective computation power to meet the needs of future devices for integration of high-throughput multimedia networks. Current state of the art in media integration technology is discussed and objectives for a new processor architecture are described. Specific parameters that contribute to the performance of multimedia adapters are identified. A set of instructions has been developed to meet the objectives, and each component of the new architecture is explained with respect to the mechanism for implementation. A single-cycle parallel multiplier provides the digital signal processing ability for voice and video processing. A frame and header processing unit provides a structure to enhance layered protocol processing and elements to facilitate encryption and data compression. Justifications for the techniques are provided and the advantages over existing technology are discussed. A brief discussion indicates how such a processor would apply to future network interfaces along with possible future trends.<<ETX>>

The effect of frequency change on discrimination of pulse bursts in an electrocutaneous tactual vocoder.

Tactual vocoders (artificial hearing systems) transduce the acoustic energy of speech into patterns of stimulation that are presented to the skin. In an electrocutaneous tactual vocoder, energy within an acoustic-frequency band is generally represented at a particular skin locus by the rate or frequency of brief electrical pulse bursts. At present little is known about tactual sensitivity for changes in the frequency of brief, bipolar electrical pulses used in several current electrocutaneous vocoder designs. Accordingly, discrimination of frequency change of electrocutaneous bipolar signals was evaluated for standard frequencies of 48, 100, 148, 200, and 248 Hz at signal durations of 50, 100, and 250 ms. Bipolar pulses (height = 10 mA, width = 13 microseconds) were presented on a single electrode placed slightly above and 8 cm to either the left or right of the navel. In a same-different task, three practiced subjects judged pairs of stimuli separated by an interstimulus interval (ISI) of 300 ms. For standard frequencies of 48 and 100 Hz, psychometric functions were similar for all subjects and all signal durations. For these frequencies, delta f/f was constant at approximately 0.3. By contrast, for standard stimuli greater than 100 Hz, the Weber fraction was found to increase dramatically as a function of both standard frequency and signal duration. In a second, similar experiment the Weber fraction for a 248-Hz standard increased as ISI was decreased below 300 ms. By contrast, ISI had little effect on sensitivity for standard stimuli of 48 and 100 Hz. Overall, these results suggest guidelines for possible intensity coding schemes for future electrocutaneous vocoder designs.

Frequency discrimination for single channel electrocutaneous signals

Frequency discrimination of electrocutaneous signals was investigated for several different base frequencies at signal durations of 50, 100, and 250 ms. Bipolar pulses (height = 10 mA, width = 13 μs) were presented on a single electrode placed 1 cm above and 8 cm to either the left or right Three practiced subjects judged pairs of stimuli (ISI = 300 ms) in a same‐different task. For signal frequencies below 160 Hz, psycho‐metric functions were similar for all subjects and for all signal durations. In this range of frequencies, Δf/f was constant at approximately 0.3. By contrast, for stimuli greater than 200 Hz, the Weber fraction was found to increase both as a function of frequency and signal duration. The results will be discussed in terms of possible intensity coding schemes for electrocutaneous vocoder designs.