Lucent Technologies Inc.

Mutualink, Inc.

Alcatel Lucent

Anand Rangaswamy Setlur

Rafid Antoon Sukkar

Michael S Wengrovitz

Joseph R Mazzarella

Derrell Lipman

Richard H Ketchum

David Lynn Thomson

Rathinavelu Chengalvarayan

Joseph L LoCicero

Malan Bhatki Gandhi

Carl Dennis Mitchell

Grzegorz Szeszko

Systems and methods of providing push-to-talk communication techniques in a multimedia conferencing session are provided. For instance, a push-to-talk mode can be implemented in a conferencing session associated with a conferencing system comprising a plurality of endpoints. At least one of the endpoints can be a full duplex endpoint. A request for transmission capabilities can be received from a first endpoint. Transmission capabilities can be granted to the first endpoint based at least in part on one or more floor control parameters associated with the conferencing session. Subsequent to granting transmission capabilities to the first endpoint, audio data can be provided from the first endpoint to at least one endpoint of the plurality of endpoints in the conferencing session.

Implementing push-to-talk in a multimedia conferencing system

Embodiments include a system, method, and computer program product that enable secure access to cameras in smart buildings. Some embodiments control outbound video from an environment such as a local network through an intelligent on-event video pushing mechanism. The local intelligent on-event video pushing mechanism hides the IP address of a source video camera, transcodes the video to a reduced size for wide area distribution, and pushes video to a recipient upon an event triggered received within the local environment (e.g., the local network.) Embodiments enable a remote video client on the far-side of the local network firewall to view the video streams of cameras on the near-side of the local network firewall when an event or trigger occurs.

System and method for secure access to camera systems

An apparatus in one example, the apparatus comprising a server configured to host a videoconference call comprising a plurality of voice and video streams, the server being communicatively associable with a plurality of clients in the videoconference call. The server being configured to determine a computing capacity of each of the plurality of clients and to hand over hosting the videoconference call to a selected one of the plurality of clients based on a computing capacity, wherein hosting the videoconference call comprises mixing the plurality of the voice and video streams into a consolidated stream, where the consolidated stream is communicated to all of the clients other than the selected client.

Offload of server-based videoconference to client-based video conference

Speech recognition technology has attained maturity such that the most likely speech recognition result has been reached and is available before an energy based termination of speech has been made. The present invention innovatively uses the rapidly available speech recognition results to provide intelligent barge-in for voice-response systems, to count words to output sub-sequences to provide paralleling and/or pipelining of tasks related to the entire word sequence, and to count words to provide rapid, speech recognition based termination of speech processing and outputting of the recognized word sequence.

Method and apparatus for word counting in continuous speech recognition useful for reliable barge-in and early end of speech detection

Echoes caused by discontinuities in a telecommunications transmission lines cause double talk which cause divergence problems with echo cancellers. Divergence problems in echo cancellers cause considerable problems in interactive caller response systems, they cause less problems if all parties to a call are human. One of the biggest problems is the effect of the echoes on automatic speech recognition systems. The echo canceller method and system presented cancels echoes to reduce their interference with ASR and yet provides stability in the presence of double talk.

Robust double-talk detection and recovery in a system for echo cancelation

A system and method provide ASR talkoff suppression in a prompt-and collect voice transaction system in which a prompt signal is sent and an input signal is received. A comparison is made of a characteristic of the prompt signal and the input signal. Processing of the input signal to reduce the likelihood of an ASR response thereto is performed if the result of the comparison satisfies a predetermined criterion.

ASR talkoff suppressor

An automatic speech recognition system for the condition that an incoming caller's speech is quiet and a resulting echo (of a loud playing prompt) can cause the residual (the portion of the echo remaining after even echo cancellation) to be of the magnitude of the incoming speech input. Such loud echoes can falsely trigger the speech recognition system and interfere with the recognition of valid input speech. An echo model has been proven to alleviate this fairly common problem and to be effective in eliminating such false triggering. Further, this automatic speech recognition system enhanced the recognition of valid speech was provided within an existing hidden Markov modeling framework.

HMM-based echo model for noise cancellation avoiding the problem of false triggers

Robust, multi-faceted sub-word method for rapidly and reliably detecting a barge-in condition of a speaker talking while an automated audio prompt is being played. This sub-word method allows for rapid stopping of the prompt to improve automatic speech recognition and reduce speaker confusion and/or frustration. An automatic speech recognition system (ASR) that practices such a method is also presented.

Method and apparatus for recognition-based barge-in detection in the context of subword-based automatic speech recognition

Method and apparatus for word counting in continuous speech recognition

A speech recognition method and apparatus which has a first stage to provide keyword hypotheses and a second stage to provide testing of those hypotheses by utterance verification. The utterance verification used has three separate models for each word: one keyword verification model, one misrecognition verification model, and one non-keyword verification model. Further, all three are developed independently of the recognizer keyword models. Because of this independence, the three verification models can be iteratively trained using existing speech data bases to jointly provide a minimum amount of verification errors.

Utterance verification using word based minimum verification error

A new speaker verification method, termed Mixture Decomposition Discrimination (MDD), and a new apparatus for using MDD are presented. MDD takes mixture component score information from a speaker independent recognizer and transmits this information while it is still decomposed as a mixture of component scores that indicate the response of the states of a Hidden Markov Model (HMM) before this information is combined into a single speaker independent recognizer parameter. MDD can be very effective in improving the performance of existing verification methods based on speaker dependent HMMs with cohort normalization because the errors of the two speaker verification methods are very uncorrelated statistically. Experimental results have shown that when MDD is incorporated into a system that also uses speaker dependent HMMs, the resulting hybrid system has its average equal error rate reduced by 46% compared to cohort normalized speaker independent HMMs. MDD is used with a speaker dependent linear discriminator which has relatively low computational and storage requirements. Thus, the increased performance of a hybrid MDD-Cohort Normalized HMM system may be achieved with minimal increase in computational and data storage assets.

Growing community of inventors

Naperville, IL, United States of America

Anand Rangaswamy Setlur