International Business Machines Corporation

Commtouch Software Ltd.

The Lithuanian University of Health Sciences

Aharon Satt

Amir Lev

Ron Hoory

Zvi Kons

Yossef Cohen

Gilad Cohen

Hagai Krupnik

Doron Hoffman

Jeffrey Haskell Derby

Tal Drory

Aya Soffer

Uzi Shvadron

Virgilijus Ulozas

A computer-implemented method, a computer system and a computer program product enhance an intent classifier through training data augmentation. The method includes selecting a target sample from a plurality of samples. The method also includes determining an ambiguity level for the target sample based on confidence scores of at least two intent labels associated with the target sample. The method further includes selecting a nearest neighboring sample from a group of neighboring samples when the ambiguity level is below a threshold. The nearest neighboring sample includes a confidence score associated with an intent label. The method also includes, for every intent label, merging the confidence scores of the two samples into an overall confidence score for the intent label and modifying the ambiguity level using the overall confidence score. Lastly, the method includes labeling the target sample with the intent label when the modified ambiguity level is above the threshold.

Intent classification enhancement through training data augmentation

A method comprising: receiving an interactive voice response (IVR) tree configured to implement one or more tasks, each associated with one or more IVR node paths comprising a plurality of IVR nodes arranged in a hierarchical relationship; analyzing the IVR tree to identify one or more intent IVR nodes, each associated with one of the tasks; with respect to each of the intent IVR nodes, identifying a plurality of corresponding entity IVR nodes included within the IVR node path associated with the intent IVR node; assembling one or more task-specific chatbot skills, each comprising (i) one of the intent IVR nodes, and (ii) at least some of the plurality of corresponding entity IVR nodes, wherein each of the task-specific chatbot skills is configured to perform one of the tasks by conducting a dialog with a user; and generating a chatbot comprising at least one of the task-specific chatbot skills.

Automated chatbot generation from an interactive voice response tree

Methods, apparatus and computer program products implement embodiments of the present invention that include receiving a time domain voice signal, and extracting a single pitch cycle from the received signal. The extracted single pitch cycle is transformed to a frequency domain, and the misclassified roots of the frequency domain are identified and corrected. Using the corrected roots, an indication of a maximum phase of the frequency domain is generated.

Vocal source extraction by maximum phase detection

Some embodiments of the invention relate to a method and a system for detecting unwanted conversational media session data. In accordance with one aspect of the invention, a method of detecting unwanted conversation media session data according to some embodiments of the invention may include calculating two or more progressive similarity scores each with respect to a different instant during a progress of a real-time conversational media session, wherein each of said scores is associated with a similarity between the conversational media session's media data that was available at the associated instant and a reference data item corresponding to media data of a previous conversational media session, and evaluating progressive similarity between the real-time conversational media session and the reference data item based upon the two or more progressive similarity scores.

Device, method and system for detecting unwanted conversational media session

Apparatus is described for digitally encoding an input audio signal for storage or transmission. A distinguishing parameter is measure from the input signal. It is determined from the measured distinguishing parameter whether the input signal contains an audio signal of a first type or a second type. First and second coders are provided for digitally encoding the input signal using first and second coding methods respectively and a switching arrangement directs, at any particular time, the generation of an output signal by encoding the input signal using either the first or second coders according to whether the input signal contains an audio signal of the first type or the second type at that time. A method for adaptively switching between transform audio coder and CELP coder, is presented. In a preferred embodiment, the method makes use of the superior performance of CELP coders for speech signal coding, while enjoying the benefits of transform coder for other audio signals. The combined coder is designed to handle both speech and music and achieve an improved quality.

Digital audio signal coding using a CELP coder and a transform coder

Apparatus is disclosed for digitally encoding an input audio signal, for storage or transmission, comprising: a pitch detector for determining at least a dominant time-domain periodicity in the input signal; a generator for generating a prediction signal based on the dominant time domain periodicity of the input signal; a first discrete frequency domain transform generator for generating a frequency domain representation of the input signal; a second discrete frequency domain transform generator for generating a frequency domain representation of the prediction signal; a subtractor to subtract at least a portion of the frequency domain representation of the prediction signal from the frequency domain representation of the input signal to generate an error signal; and a generator to generate an output signal from the error signal and parameters defining the prediction signal. A corresponding decoder is also described.

Digital audio signal coding

A time-division multiplexing synchronization mechanism employs a 1-bit/frame framing channel with a pattern that repeats every 2 frames. Acquisition and reacquisition of synchronization with such a framing pattern is enabled through the use of an explicit synchronization procedure. With this procedure, initial acquisition of frame synchronization as well as reframing after loss of synchronization are carried out while the transmission channel is carrying only a predefined bit pattern. When the receivers are in synchronization, synchronization is monitored and maintained using a 1-bit/frame framing channel, while the communications link is carrying multiplexed bit streams with the appropriate format.

Growing community of inventors

Haifa, Israel

Aharon Satt