Abstract: A “speech onset detector” provides a variable length frame buffer in combination with either variable transmission rate or temporal speech compression for buffered signal frames. The variable length buffer buffers frames that are not clearly identified as either speech or non-speech frames during an initial analysis. Buffering of signal frames continues until a current frame is identified as either speech or non-speech. If the current frame is identified as non-speech, buffered frames are encoded as non-speech frames. However, if the current frame is identified as a speech frame, buffered frames are searched for the actual onset point of the speech. Once that onset point is identified, the signal is either transmitted in a burst, or a time-scale modification of the buffered signal is applied for compressing buffered frames beginning with the frame in which onset point is detected. The compressed frames are then encoded as one or more speech frames.

Abstract: A multi-sensor sound source localization (SSL) technique is presented which provides a true maximum likelihood (ML) treatment for microphone arrays having more than one pair of audio sensors. Generally, this is accomplished by selecting a sound source location that results in a time of propagation from the sound source to the audio sensors of the array, which maximizes a likelihood of simultaneously producing audio sensor output signals inputted from all the sensors in the array. The likelihood includes a unique term that estimates an unknown audio sensor response to the source signal for each of the sensors in the array.

Abstract: A system and method for embedding information into digital media and later detecting the embedded information using a unique spread spectrum modulation technique. In general, the present invention removes interference caused by an original signal from the detection process thereby eliminating a major source of detection error. The interference caused by the original signal is removed by using the encoder knowledge about the original signal and modulating the energy of the embedded mark to compensate for the original signal interference. The present invention also includes a novel redundant bit representation technique causes a resulting average over a large sample to tend to zero, thereby reducing the vulnerability of the present invention to malicious collusion attacks.

Abstract: An adaptive “temporal audio scaler” is provided for automatically stretching and compressing frames of audio signals received across a packet-based network. Prior to stretching or compressing segments of a current frame, the temporal audio scaler first computes a pitch period for each frame for sizing signal templates used for matching operations in stretching and compressing segments. Further, the temporal audio scaler also determines the type or types of segments comprising each frame. These segment types include “voiced” segments, “unvoiced” segments, and “mixed” segments which include both voiced and unvoiced portions. The stretching or compression methods applied to segments of each frame are then dependent upon the type of segments comprising each frame. Further, the amount of stretching and compression applied to particular segments is automatically variable for minimizing signal artifacts while still ensuring that an overall target stretching or compression ratio is maintained for each frame.

Abstract: Confidential information is provided to a proxy computer in communication between an unsecured computer and a computer having information desired by a user. The proxy computer receives the confidential information in either an encrypted form or having arbitrary information combined therewith. The proxy computer ascertains the confidential information and forwards it to the computer having the information desired by the user.

Abstract: An acoustic echo cancellation technique. The present adaptive acoustic echo cancellation technique employs a plurality of acoustic echo cancellation filters which use different adaptation techniques which may employ different parameters such as step size, to improve both the adaptation algorithm convergence time and misadjustment over previously known acoustic echo cancellation techniques.

Abstract: A unique system and method that facilitate automatic tracking of user data and providing on-demand site or merchant reputation information is provided. The system and method involve monitoring user input such as by keystrokes over a period of time and forming a long continuous string of characters. The string of characters can be monitored for repeating substrings or non-repeating substrings that indicate sensitive user information such as SSN or credit card number (CCN). The substrings can be displayed to the user in a digest form and order according to frequency of occurrence, for example. Thus, the user can readily view where and when their sensitive information has been entered. In addition, the typing of certain substrings such as CCNs or other sensitive field data can trigger a reputation service to retrieve the reputation data for the current site or merchant to mitigate dealing with fraudulent sites or merchants.

Abstract: A decentralized computer network architecture and method that gathers metadata from local and remote clients and, based on that metadata, locally makes a decision whether to send a packet over the network. Each client listens to what other clients are doing, and only sends when the total number of concurrent speakers is below some threshold. In a multi-party voice conferencing embodiment, the threshold is a number of concurrent speakers that is restricted to less than a certain number. Under the decentralized computer network architecture, the type of network topology used to connect the clients is flexible, as long as each client is running a peer-aware system to decide locally whether to send their packets. The decentralized computer network architecture and method is distributed to run on each client, making it suitable for a wide variety of network topologies (such as full-mesh, bridge-based, or a hybrid of the two).

Abstract: A peer-aware voice stream ranking method that makes decisions based on information about participants of a voice conference over a network. Whether to send a participant's own audio packet out on the network is based both on information about the participant's own voice packet and voice packets that the participant receives from other clients. A Voice Activity Score (VAS) is computed for each frame of a particular voice stream. The VAS includes a voiceness component, indicating the likelihood that the audio frame contains speech or voice, and an energy level component that indicating the ratio of current frame energy to the long-term average of energy for a current speaker. Using the VAS from the participants, the method also ranks the client's voice stream as compared to other clients' voice streams in the voice conference. If there are participants higher ranking, the client's voice stream is not sent.

Abstract: A phishing detection client component and method is provided. The component can be employed as part of a system to detect and, optionally, prevent phishing attacks. The phishing detection client component can provide password reuse event report(s), for example, to a phishing detection server component. The client component can further include a credential component that can track use of credentials by a user and determine whether a specific security credential is being used or presented. Due to the malicious nature of phishing in general, the client component can be susceptible to attacks by phishers. For example, phishers can generate false logins in an attempt to flood the client component with information resulting in induced false positives and/or induced false negatives. The client component can perform one or more checks to determine whether false login(s) have been attempted.

Abstract: A system and process for finding the location of a sound source using direct approaches having weighting factors that mitigate the effect of both correlated and reverberation noise is presented. When more than two microphones are used, the traditional time-delay-of-arrival (TDOA) based sound source localization (SSL) approach involves two steps. The first step computes TDOA for each microphone pair, and the second step combines these estimates. This two-step process discards relevant information in the first step, thus degrading the SSL accuracy and robustness. In the present invention, direct, one-step, approaches are employed. Namely, a one-step TDOA SSL approach and a steered beam (SB) SSL approach are employed. Each of these approaches provides an accuracy and robustness not available with the traditional two-step approaches.

Abstract: A phishing detection server component and method is provided. The component can be employed as part of a system to detect/phishing attacks. The phishing detection server component can receive password reuse event report(s), for example, from a protection component of client component(s). Due to the malicious nature of phishing in general, the phishing detection server component can be susceptible to attacks by phishers (e.g., by reverse engineering of the client component). For example, false report(s) of PREs can be received from phisher(s) in an attempt to overwhelm the server component, induce false positives and/or induce false negatives. Upon receipt of a PRE report, the phishing detection server component can first verify that the timestamp(s) are genuine (e.g., previously generated by the phishing detection server component). The report verification component can employ the timestamp(s) to verify veracity of the report (e.g., to minimize attacks by phishers).

Abstract: A system for protecting against information security breaches comprises a credential module that maintains a list of protected security credentials that are each associated with a known computing system and that detects when a security credential in the list is used. The system also includes a protection module that detects use of the security credential in connection with a computing system other than the known computing system with which the security credential is associated and that reports such use to a security server. Methods for using the system are also provided.

Abstract: A method and system for modifying a digital audio signal to vary its playback speed while preserving the signal's pitch and quality. The variable speed playback (VSP) system and method mitigates artifacts remaining after processing by existing techniques. The VSP system and method produces a consistent and pleasing sound to an audio file, even while its speed is varied during playback. The VSP method includes selecting and estimating an input frame, adjusting the frame position, and overlapping and adding the adjust frame to an output signal. The frame position adjustment is achieved using an enhanced correlation technique that finds all local maxima over a cross-correlation function. The local maxima having a highest correlation score is designated as a cut position, where the adjusted frame is cut from the input buffer. The VSP system and method using four input frames to generate one output frame.

Abstract: A system and process for finding the location of a sound source using direct approaches having weighting factors that mitigate the effect of both correlated and reverberation noise is presented. When more than two microphones are used, the traditional time-delay-of-arrival (TDOA) based sound source localization (SSL) approach involves two steps. The first step computes TDOA for each microphone pair, and the second step combines these estimates. This two-step process discards relevant information in the first step, thus degrading the SSL accuracy and robustness. In the present invention, direct, one-step, approaches are employed. Namely, a one-step TDOA SSL approach and a steered beam (SB) SSL approach are employed. Each of these approaches provides an accuracy and robustness not available with the traditional two-step approaches.

Abstract: A system and process for finding the location of a sound source using direct approaches having weighting factors that mitigate the effect of both correlated and reverberation noise is presented. When more than two microphones are used, the traditional time-delay-of-arrival (TDOA) based sound source localization (SSL) approach involves two steps. The first step computes TDOA for each microphone pair, and the second step combines these estimates. This two-step process discards relevant information in the first step, thus degrading the SSL accuracy and robustness. In the present invention, direct, one-step, approaches are employed. Namely, a one-step TDOA SSL approach and a steered beam (SB) SSL approach are employed. Each of these approaches provides an accuracy and robustness not available with the traditional two-step approaches.

Abstract: A system and process for finding the location of a sound source using direct approaches having weighting factors that mitigate the effect of both correlated and reverberation noise is presented. When more than two microphones are used, the traditional time-delay-of-arrival (TDOA) based sound source localization (SSL) approach involves two steps. The first step computes TDOA for each microphone pair, and the second step combines these estimates. This two-step process discards relevant information in the first step, thus degrading the SSL accuracy and robustness. In the present invention, direct, one-step, approaches are employed. Namely, a one-step TDOA SSL approach and a steered beam (SB) SSL approach are employed. Each of these approaches provides an accuracy and robustness not available with the traditional two-step approaches.

Abstract: A system and process for estimating the time delay of arrival (TDOA) between a pair of audio sensors of a microphone array is presented. Generally, a generalized cross-correlation (GCC) technique is employed. However, this technique is improved to include provisions for both reducing the influence (including interference) from correlated ambient noise and reverberation noise in the sensor signals prior to computing the TDOA estimate. Two unique correlated ambient noise reduction procedures are also proposed. One involves the application of Wiener filtering, and the other a combination of Wiener filtering with a Gnn subtraction technique. In addition, two unique reverberation noise reduction procedures are proposed. Both involve applying a weighting factor to the signals prior to computing the TDOA which combines the effects of a traditional maximum likelihood (TML) weighting function and a phase transformation (PHAT) weighting function.

Abstract: Real-time packet-based audio communications over packet-based networks frequently results in the loss of one or more packets during any given communication session. The real-time nature of such communications precludes retransmission of lost packets due to the unacceptable delays that would result. Consequently, packet loss concealment methods are employed to “hide” lost packets from the listener. Unfortunately, conventional loss concealment methods, such as packet repetition or stretch/overlap methods, do not fully exploit information available from partially received samples. Therefore, when a single frame of N coefficients is lost, 2N samples are only partially reconstructed, thereby degrading the reconstructed signal.

Abstract: A system and method facilitating signal enhancement utilizing an adaptive filter is provided. The invention includes an adaptive filter that filters an input based upon a plurality of adaptive coefficients and modifies the adaptive coefficients based on a feedback output. A feedback component provides the feedback output based, at least in part, upon a non-linear function of the acoustic reverberation reduced output. Optionally, the system can further include a linear prediction (LP) analyzer and/or a LP synthesis filter. The system can enhance signal(s), for example, to improve the quality of speech that is acquired by a microphone by reducing reverberation. The system utilizes, at least in part, the principle that certain characteristics of reverberated speech are measurably different from corresponding characteristics of clean speech. The system can employ a filter technology (e.g., reverberation reducing) based on a non-linear function, for example, the kurtosis metric.