Abstract: A system and method are presented for advanced campaign management in outbound dialing platforms. In a contact center environment, a dialing campaign may be configured such that the outbound dialing platform modifies its behavior automatically throughout the dialing campaign as conditions change. Dialing campaigns may be constructed from groups, which are organized into a campaign sequence. As the campaign sequence flows between groups, details of the campaign may be monitored and evaluated by the platform. Group transitions may be initiated based on activation triggers. In an embodiment, campaign sequences may also be automatically transitioned to new campaign sequences, paused, or reset based on activation triggers. A transition may also be initiated by manual override of the platform.

Abstract: A system and method are presented for neural network based feature extraction for acoustic model development. A neural network may be used to extract acoustic features from raw MFCCs or the spectrum, which are then used for training acoustic models for speech recognition systems. Feature extraction may be performed by optimizing a cost function used in linear discriminant analysis. General non-linear functions generated by the neural network are used for feature extraction. The transformation may be performed using a cost function from linear discriminant analysis methods which perform linear operations on the MFCCs and generate lower dimensional features for speech recognition. The extracted acoustic features may then be used for training acoustic models for speech recognition systems.

Abstract: A system and method are presented for acoustic data selection of a particular quality for training the parameters of an acoustic model, such as a Hidden Markov Model and Gaussian Mixture Model, for example, in automatic speech recognition systems in the speech analytics field. A raw acoustic model may be trained using a given speech corpus and maximum likelihood criteria. A series of operations are performed, such as a forced Viterbi-alignment, calculations of likelihood scores, and phoneme recognition, for example, to form a subset corpus of training data. During the process, audio files of a quality that does not meet a criterion, such as poor quality audio files, may be automatically rejected from the corpus. The subset may then be used to train a new acoustic model.

Abstract: Systems and methods for the matching of datasets, such as input audio segments, with known datasets in a database are disclosed. In an illustrative embodiment, the use of the presently disclosed systems and methods is described in conjunction with recognizing known network message recordings encountered during an outbound telephone call. The methodologies include creation of a ternary fingerprint bitmap to make the comparison process more efficient. Also disclosed are automated methodologies for creating the database of known datasets from a larger collection of datasets.

Abstract: A system and method are presented for the synthesis of speech from provided text. Particularly, the generation of parameters within the system is performed as a continuous approximation in order to mimic the natural flow of speech as opposed to a step-wise approximation of the feature stream. Provided text may be partitioned and parameters generated using a speech model. The generated parameters from the speech model may then be used in a post-processing step to obtain a new set of parameters for application in speech synthesis.

Abstract: A method and system for prediction in a telecommunications network are described. Particularly, contact allocation, staff time distribution, and other performance metrics may be determined in a contact center operation environment comprising multiple media types and/or multiple agent skill sets for handling contact types. In an embodiment, models simulating queue behavior are built. Staff count may be altered in order to determine staffing allocations and performance scenarios using the models. Long term and short term predictions may be used to configure the telecommunication network and route communications through the telecommunications network.

Abstract: A system and method are presented for the use of natural language processing to automate the disposition of completed interactions in an environment such as a contact center system. Communications may be inbound or outbound. A communication is received or placed by the contact center system and processed through the contact center system. After the interaction of the communication and the contact center is completed, data from the completed interaction may be processed through an automated customer interaction disposition system. Transcript data from the interaction may be processed through an NLP engine to generate data from the interaction. The data may then be automatically matched with a disposition/wrap-up code for the interaction.

Abstract: A system and method for learning alternate pronunciations for speech recognition is disclosed. Alternative name pronunciations may be covered, through pronunciation learning, that have not been previously covered in a general pronunciation dictionary. In an embodiment, the detection of phone-level and syllable-level mispronunciations in words and sentences may be based on acoustic models trained by Hidden Markov Models. Mispronunciations may be detected by comparing the likelihood of the potential state of the targeting pronunciation unit with a pre-determined threshold through a series of tests. It is also within the scope of an embodiment to detect accents.

Abstract: Systems and methods for the matching of datasets, such as input audio segments, with known datasets in a database are disclosed. In an illustrative embodiment, the use of the presently disclosed systems and methods is described in conjunction with recognizing known network message recordings encountered during an outbound telephone call. The methodologies include creation of a ternary fingerprint bitmap to make the comparison process more efficient. Also disclosed are automated methodologies for creating the database of known datasets from a larger collection of datasets.

Abstract: A system and method are presented for selectively biased linear discriminant analysis in automatic speech recognition systems. Linear Discriminant Analysis (LDA) may be used to improve the discrimination between the hidden Markov model (HMM) tied-states in the acoustic feature space. The between-class and within-class covariance matrices may be biased based on the observed recognition errors of the tied-states, such as shared HMM states of the context dependent tri-phone acoustic model. The recognition errors may be obtained from a trained maximum-likelihood acoustic model utilizing the tied-states which may then be used as classes in the analysis.

Abstract: A system and method are presented for real-time speech analytics in the speech analytics field. Real time audio is fed along with a keyword model, into a recognition engine. The recognition engine computes the probability of the audio stream data matching keywords in the keyword model. The probability is compared to a threshold where the system determines if the probability is indicative of whether or not the keyword has been spotted. Empirical metrics are computed and any false alarms are identified and rejected. The keyword may be reported as found when it is deemed not to be a false alarm and passes the threshold for detection.

Abstract: A system and method are presented for analysis of a connection comprising a communication provider and a remote device. Known timing conditions may be used in the analysis to eliminate the impact of analysis delay on the contacted party flow. A point in time may be determined at which a live answer is no longer expected and an answering service answer may be expected using several factors. These factors comprise the routing time to/from the originating provider to a secondary provider as well as the time that the secondary provider alerts before transferring to a voice message service. Live answer determination may be categorized prior to the indication of a far-end connection and before the application of any media analysis routines.

Abstract: A system and method are presented for automatic communication distribution local assignment. An on-premises device, in a contact center environment using a cloud-based solution, communicates with the solution to determine if there is a service interruption. When an interruption is determined, the on-premises device steps in to perform local assignment of communications for the solution until communication has been restored with the cloud-based solution. The on-premises device comprises a module that monitors all ACD communication flows. The module may also manage the logic of when to intervene and how to assign communications in queue to available agents during outages.

Abstract: A system and method are presented for using spoken word verification to reduce false alarms by exploiting global and local contexts on a lexical level, a phoneme level, and on an acoustical level. The reduction of false alarms may occur through a process that determines whether a word has been detected or if it is a false alarm. Training examples are used to generate models of internal and external contexts which are compared to test word examples. The word may be accepted or rejected based on comparison results. Comparison may be performed either at the end of the process or at multiple steps of the process to determine whether the word is rejected.

Abstract: A system and method are presented for acoustic echo cancellation. The echo canceller performs reduction of acoustic and hybrid echoes which may arise in a situation such as a long-distance conference call with multiple speakers in varying environments, for example. Echo cancellation, in at least one embodiment, may be based on similarity measurement, statistical determination of echo cancellation parameters from historical values, frequency domain operation, double talk detection, packet loss detection, signal detection, and noise subtraction.

Abstract: A system and method are presented for contact identification dissemination among devices within a hybrid cloud distributed system. Information associated with a first party device is provided to a cloud managed telephony network appliance. The appliance communicates with a cloud application and is capable of communication with any other appliances which may be associated with a second party device, if that appliance is not already associated with the second party device. Information between the components in the system may be asynchronously updated in order to provide first party information and more granular matches.

Abstract: A system and method are presented for learning call analysis. Audio fingerprinting may be employed to identify audio recordings that answer communications. In one embodiment, the system may generate a fingerprint of a candidate audio stream and compare it against known fingerprints within a database. The system may also search for a speech-like signal to determine if the end point contains a known audio recording. If a known audio recording is not encountered, a fingerprint may be computed for the contact and the communication routed to a human for handling. An indication may be made as to if the call is indeed an audio recording. The associated information may be saved and used for future identification purposes.

Abstract: Methods for creating, arranging, and leveraging an ad-hoc collection of heterogeneous organization components are provided. In one example, a method includes the steps of displaying a scratchpad affordance associated with a scratchpad, and detecting an update to the scratchpad based on an input associated with a first organization component of the one or more organization components and the scratchpad affordance. The method includes the additional step of updating the scratchpad based at least in part on the input.

Abstract: Methods for a secret supplemental username are provided. In one example, a method includes the steps of detecting a threat to an account with a primary username and password and disabling the primary username associated with the account in response to the threat. The method includes the additional steps of generating a supplemental username associated with the account and securely communicating the supplemental username.

Abstract: A system and method are presented for predictive live interaction offering and hosting. A mechanism may be provided for controlling outstanding offerings, or invitations, of communications with users through points of service. Offerings may include a live chat or a telephone callback request. In one embodiment, offerings may be optimized by analyzing factors such as agent history to predict when and how many agents will be available. Calculations may be continuously performed as invitations are accepted, declined, timed out, etc., to make certain more offers than can be handled are not extended.