Abstract: Detecting a replay attack on a voice biometrics system comprises receiving a speech signal; forming an autocorrelation of at least a part of the speech signal; and identifying that the received speech signal may result from a replay attack based on said autocorrelation. Identifying that the received speech signal may result from a replay attack may be achieved by: comparing the autocorrelation with a reference value; and identifying that the received speech signal may result from a replay attack based on a result of the comparison of the autocorrelation with the reference value, or by: supplying the autocorrelation to a neural network trained to distinguish autocorrelations formed from speech signals resulting from replay attacks from autocorrelations formed from speech signals not resulting from replay attacks.

Abstract: Embodiments of the present disclosure disclose a method and apparatus for generating a text-to-speech model. A specific implementation of the method includes: obtaining a training sample set, a training sample including sample text information, sample audio data corresponding to the sample text information, and a fundamental frequency of the sample audio data; obtaining an initial deep neural network; and using the sample text information of the training sample in the training sample set as an input, and using the sample audio data corresponding to the input sample text information and the fundamental frequency of the sample audio data as an output, to train the initial deep neural network using a machine learning method, and defining the trained initial deep neural network as the text-to-speech model.

Abstract: This disclosure relates to configuration tools for interactive agents, sometimes referred to as bots, chatbots, virtual robots, or talkbots. Such interactive agents utilize slots for organizing and storing data received as inputs and displayed as outputs. These slots can be configured such that a slot is temporary and does not persist beyond its source dialog. Slots can also be configured such that a slot is pre-populated with information contain in incoming passed parameters.

Abstract: A data processing device includes: a digital signal processor; at least one processor; and at least one memory device configured to store a plurality of instructions, which when executed by the at least one processor, cause the at least one processor to operate to: output a first determination result relating to a scene of content through use of sound data; select processing for the sound data by a first selection method based on the first determination result; determine an attribute of the content from among a plurality of attribute candidates; and select the processing by a second selection method, which is different from the first selection method, based on a determination result of the attribute, wherein the digital signal processor is configured to execute the processing selected by the at least one processor on the sound data.

Abstract: In one embodiment, a computer program product includes a computer readable storage medium having program instructions embodied therewith. The program instructions are executable by a processing circuit to cause the processing circuit to receive collected data from one or more data collection devices, the collected data being aggregated over at least one month and comprising audio data including first voice input of a user of the one or more data collection devices. The program instructions also cause the processing circuit to store the audio data to a computer readable storage medium; determine an identity of the user based on comparing second voice input to the first voice input; and analyze the audio data for indications of hearing loss in the user over the period of time. The analysis includes determining a user's emotion during production of the audio input.

Abstract: A system and apparatus for recording and archiving diverse communications over radio transmissions. The system and apparatus enables unattended airports within a geofenced area to generate a useful archive of all radio communications made by Automatic Dependent Surveillance-Broadcast (ADS-B) equipped aircraft and ground personnel. A combination of hardware and software components are provided to record and store radio transmissions in computer files. Once stored, the computer files may then be replayed for training and investigation purposes. Likewise, users may generate custom reports based upon the data embodied in the computer files.

Abstract: Traditional audio feedback elimination systems may attempt to reduce the effect of the audio feedback by simply scaling down the audio volume of the signal frequencies that are prone to howling. Other traditional feedback elimination systems may also employ adaptive notch filtering to detect and “notch” the so-called “singing” or “howling” frequencies as they occur in real-time. Such devices may typically have several knobs and buttons needing tuning, for example: the number of adaptive parametric equalizers (PEQs) versus fixed PEQs; attack and decay timers; and/or PEQ bandwidth. Rather than removing the singing frequencies with PEQs, the devices described herein attempt to holistically model the feedback audio and then remove the entire feedback signal. Two advantages of the devices described herein are: 1.) the system can operate at a much larger loop-gain (and hence with a much higher loudspeaker volume); and 2) setup is greatly simplified (i.e., no tuning knobs or buttons).

Abstract: A user interface is presented for a hearing-impaired user to make selections that influence how that user participates in a video conference and how other participants in that video conference interact with the hearing-impaired user. For example, through appropriate user interface selections, the hearing-impaired user may specify that he/she is to always preview (for editing purposes, etc.) any sign language translations made by a video conferencing system before those translations are released to the other participants, whether as text and/or speech. Through other user interface selections, the user may configure linguistic and/or playback characteristics associated with the sign-language-to-speech translations, with speech signals being produced so as to include certain effects when played out at the endpoints of the other participants and/or emulate a desirable video conference persona from the standpoint of the hearing-impaired user.

Abstract: There is provided an audio encoding apparatus including a memory, and a processor coupled to the memory and the processor configured to determine whether a tone is included in a boundary between a low-frequency that is a frequency bandwidth below a predetermined frequency of an input signal and a high-frequency that is a frequency bandwidth above the predetermined frequency of the input signal, suppress a tone in one of the low-frequency and the high-frequency, encode the input signal having the low-frequency to generate a low-frequency code, encode the input signal having the high-frequency to generate a high-frequency code, and generate an encoded stream by multiplexing the low-frequency code and the high-frequency code.

Abstract: A method for generating a signature is disclosed. As part of the method, a first number of bits are identified in respective rows of an audio signature matrix that are determined to be the strongest bits in the row, bits of the audio signature matrix are replaced with respective cells having values depending on whether the respective bits are included in the first number of bits, a set of uniformly distributed numbers are generated within a range of numbered locations corresponding to cells of the audio signature matrix; numerical distances are determined, from respective numbers in the set of uniformly distributed numbers, to the numbered locations of the matrix, associated with nearest occurrences of a first value. A set of integers is generated based on the distances.

Abstract: A car audio output control device and a method therefor are disclosed. When an acoustic source is generated by the peripheral device, the control device controls a car audio output such that when a device acoustic source and an identification code are received while a source acoustic source self-generated by a car audio is outputted, the control device selects one output mode matched with the received identification code from among the plurality of output modes and controls the source acoustic source and the device acoustic source according to the selected output mode so as to change the car audio output.

Abstract: Methods and apparatus for audio identification during a performance are disclosed herein. An example method includes transforming a segment of audio into a log-frequency spectrogram based on a transform using a logarithmic frequency resolution. The log-frequency spectrogram is transformed into a binary image, each pixel of the binary image corresponding to a time frame and frequency channel pair. A matrix product of the binary image and a plurality of reference fingerprints is computed. Based on a number of frequency ranges represented in the binary image, the matrix product to form a similarity matrix is computed. Selecting an alignment of a line in the similarity matrix that intersects one or more bins in the similarity matrix with the largest calculated Hamming similarities. A reference fingerprint is selected based on the alignment.

Abstract: Embodiments for extraction and summarization of decision discussions of a communication by a processor. The decision elements may be grouped together according to similar characteristics. The decision elements may be linked, and sentiments of the discussion participants towards each of the decision elements may be analyzed. A summary of the plurality of the decision elements may be provided via an interactive graphical user interface (GUI) on one or more Internet of Things (IoT) devices. The summary of the decision elements may be linked to domain knowledge. The summary may be enhanced using a domain knowledge.

Abstract: Apparatus, systems, articles of manufacture, and methods are disclosed for multiple scrambled layers for audio watermarking. An example system includes a scrambler executing instructions to: divide a watermark into a plurality of watermark symbols; map the watermark symbols to a plurality of frequency bins in a plurality of frequency clumps according to a first distribution scheme to create a first watermark layer having a first combination of the frequency bins and a second watermark layer having a second combination of the frequency bins, the first combination of the frequency bins and the second combination of the frequency bins partially overlap. The scrambler also is to determine a sequence for shifting watermark symbols among the frequency bins, and generate a second distribution scheme to map the watermark symbols in accordance with the sequence. The example system also includes a transceiver to communicate the second distribution scheme to a device.

Abstract: A software input string is received and tokenized into a sequence of tokens. The sequence of tokens is applied to a trained sequence-dependent lock/unlock classifier so that each of the tokens is classified as a token that should be locked, or remain unlocked, for subsequent translation. The software input string is converted to a converted string, in which the locked tokens are identified and the converted string is submitted for machine translation. A machine translation result is received and converted so that the locked tokens are replaced in the machine translation result, to obtain a translated software string.

Abstract: In one embodiment, an apparatus comprising a first audio looping device is provided. The first audio looping device is electrically coupled to a computing device. The first audio looping device is programmed to receive a first audio signal from a musical instrument and to store the first audio signal. The first audio looping device is further programmed to playback the stored first audio signal as a first audio loop a number of times and to transmit the first audio loop to a second audio looping device via the first computing device. The first audio looping device is further programmed to receive a second audio loop from the second audio looping device via the computing device.

Abstract: Apparatuses, methods and a storage medium associated with a sensor hub are disclosed herein. In embodiments, a sensor hub may include a body with a plurality of apertures disposed on the sensor hub body at substantially equal distances to each other, oriented towards a plurality of directions, and a plurality of sensors disposed inside the sensor hub body to interface respective apertures, to sense ambient conditions in the plurality of directions via the apertures. The sensor hub may further include one or more communication interfaces to transmit the sensor data or results from processing the sensor data to an external recipient. Other embodiments may be described and claimed.

Abstract: The present disclosure discloses an artificial intelligence based method and apparatus for segmenting a sentence. A specific embodiment of the method includes: lexing a to-be-segmented original sentence to obtain a set of words in the original sentence; performing sentence segmentation steps on a to-be-segmented sentence having an initial value being the original sentence; using the sub-sentence fragment not belonging to the set of words as the to-be-segmented sentence and continuing to perform the sentence segmentation steps, if the sub-sentence fragment not belonging to the set of words exists; and storing the original sentence in association with the plurality of sub-sentence fragments each time obtained by performing the sentence segmentation steps. The embodiment generates a segmentation result obtained by performing a multi-layered segmentation to the original sentence.

Abstract: An analysis device includes a memory and a processor configured to acquire a document for analyzing, generate a coded document including a plurality of coded words by coding the document for each word unit, determine a vector value according to a condition of appearance of a first coded phrase in the coded document, the first coded phrase corresponding to a combination of two or more coded words included in the plurality of coded words, and assign the vector value to the first coded phrase.

Abstract: The present invention relates to a method and an apparatus for processing a signal, which are used to effectively reproduce an audio signal, and more particularly, to a method for generating a filter for an audio signal, which are used for implementing a filtering for input audio signals with a low computational complexity and a parameterization apparatus therefor.