Abstract: An electrophoretic display device and a method for electrophoretic display refreshing are provided in the disclosure. The electrophoretic display device includes a display screen, a driving circuit, a driving control circuit, and a processor. The display screen includes multiple pixel units. The driving circuit is configured to generate multiple first data-voltage signals at a first time and multiple second data-voltage signals at a second time. The processor is configured to obtain and compare one of the multiple first data-voltage signals and one of the multiple second data-voltage signals that both are to be received by a same pixel unit of the multiple pixel units, to obtain a comparison result, and the processor is configured to control, according to the comparison result, the driving circuit and the driving control circuit to drive the pixel unit to work.

Abstract: Methods for driving an electrophoretic medium including two pairs of oppositely charged particles. The first pair including a first type of positive particles and a first type of negative particles and the second pair consists of a second type of positive particles and a second type of negative particles, wherein the first pair of particles and the second pair of particles have different charge magnitudes (identifiable as zeta potentials). In particular, the driving methods produce cleaner optical stakes of the lesser-charged particles with less contamination from the other particles and more consistent electro-optical performance when the intermediate driving voltages are modified.

Abstract: It is an object to provide a semiconductor device which can supply a signal with sufficient amplitude to a scan line while power consumption is kept small. Further, it is an object to provide a semiconductor device which can suppress distortion of a signal supplied to the scan line and shorten a rising time and a falling time while power consumption is kept small. A semiconductor device which includes a plurality of pixels each including a display element and at least one first transistor and a scan line driver circuit supplying a signal for selecting the plurality of pixels to a scan line. A light-transmitting conductive layer is used for a pixel electrode layer of the display element, a gate electrode layer of the first transistor, source and drain electrode layers of the first transistor, and the scan line.

Abstract: A power-optimized monitor includes a universal serial bus (USB) port for receiving power from a host computing device. The monitor is configured to be powered on 2.5 Watts for USB 2.0 standard and 4.5 Watts for USB 3.0 standard. Further, a method includes receiving power from the host computing device with a universal serial bus (USB) port of the monitor; displaying, with a display screen of the monitor, video images contained in the digital video signal received from the host computing device; decoding, with an input converter of the monitor, the digital video signal at the single input port from a first interface standard to a second interface standard; encoding, with an output converter of the monitor, the digital video signal in the second interface standard to a third interface standard, which matches an interface standard of the display screen.

Abstract: A display system supports variable refresh rates that include a plurality of refresh rates. A source such as a graphics processing unit (GPU) provides frames to the display system at a selected one of the refresh rates. The refresh rates are factored into a corresponding plurality of prime factors. A plurality of numbers of lines per frame in frames provided at the plurality of refresh rates is determined based on one or more ratios of the plurality of refresh rates, the plurality of prime factors, and a line rate for providing frames to the display system at the plurality of refresh rates. The source then selectively provides frames to the display system at one refresh rate of the plurality of refresh rates using the same line rate regardless of which refresh rate is chosen. Furthermore, the number of lines per frame is an integer for frames provided at the refresh rates.

Abstract: There is provided an information processing device including a voltage detection unit configured to monitor a voltage value of a signal output at a predetermined timing, and a signal control unit configured to stop output of the signal if the voltage value after a predetermined time elapses from when the voltage value detected by the voltage detection unit exceeds a first value does not exceed a second value greater than the first value.

Abstract: A method includes obtaining rendered image data that includes a representation of an object for display using a see-through display. The see-through display permits ambient light from a physical environment through the see-through display. The method includes sensing a plurality of light superposition characteristic values associated with the ambient light that quantifies the ambient light. The method includes determining a plurality of display correction values associated with the electronic device based on the plurality of light superposition characteristic values and predetermined display characteristics of the representation of the object. The method includes generating, from the rendered image data, display data for the see-through display in accordance with the plurality of display correction values in order to satisfy the predetermined display characteristics of the representation of the object within a performance threshold.

Abstract: A system and method for synchronizing two or more display elements of a multiple element display, is disclosed. The method includes embedding a frame count signal into the blanking portion of a video frame for each data stream received by the multiple display elements. Upon displaying the pixel data from a video frame, a frame count signal containing an identifying frame count is transmitted back to the computer. The computer compares the frame count signals to determine if the multiple display elements are synchronized. Upon a determination that the multiple display elements are not synchronized, the data stream for one or more display elements are adjusted accordingly. The system may be configurated as a federated system with two or more computers each communicatively coupled to each display element in a hierarchal system, where a secondary computer can control a display element if the primary computer or primary computer link fails.

Abstract: A hybrid-material construction for string instruments may include providing an internal lattice framework structure encased in a composite material to produce a final form of a string instrument. The framework structure may provide a lightweight structural support system within a finished instrument to reduce overall weight while maintaining strength of the instrument.

Abstract: A sound processing system includes an electronic musical instrument and a sound processing apparatus communicable with the instrument. The instrument includes an audio signal generator that generates an audio signal according to a user performance on the electronic musical instrument, a first signal processor that performs first effect processing on the audio signal to generate a first processed audio signal, a first sound emitter that emits a first performance sound component based on at least one of the first processed audio signal or a second processed audio signal, and a first audio signal output that outputs the audio signal.

Abstract: A computer-implemented sound processing method includes: outputting singing sound data based on a sound signal representing singing sound; and outputting sound data representing musical instrument sound that correlates with musical elements of the singing sound, by inputting input data that includes the singing sound data to a trained model that has learned, by machine learning, a relationship between singing sound for training and musical instrument sound for training.

Abstract: An acoustic absorbing filler, the acoustic absorbing filler comprising a core particle comprising a polymer; an outer layer coated on the core particle, wherein the outer layer comprise microporous particulates; and wherein the acoustic absorbing filler has a median particle size of from 100 micrometer to 700 micrometers and a specific surface area of from 10 m2/g to 400 m2/g; wherein the acoustic absorbing filler has a normal incidence acoustic absorption of no less than 0.15 at 300 Hz when measured in a 20 mm packed bed.

Abstract: A system including a noise analyzer, a filter, and a memory. The noise analyzer configured to review a captured audio signal, wherein the captured audio signal comprises baseline noise parameters and clean audio. The filter configured to remove the baseline noise parameters from the captured audio signal so that the clean audio is free of the baseline noise parameters. The memory stores the clean audio.

Abstract: An active noise reduction system includes a canceling sound generator configured to generate a canceling sound for canceling a noise, an error detector configured to detect an error between the noise and the canceling sound and generate an error signal corresponding to the error, and a controller configured to control the canceling sound generator based on the error signal, wherein the controller is configured to extract noise components at a plurality of frequencies based on the error signal, determine a control target frequency among the plurality of frequencies based on the noise components at the plurality of frequencies, select a value of a prescribed control parameter based on the control target frequency, and generate a control signal to control the canceling sound generator based on the selected value of the control parameter.

Abstract: An acoustic signal cancellation system is configured to perform: (a) provide a user device, associated with the system, and an audio capture device. (b) receive at system a captured signal indicative of a captured acoustic signal. It's received at capture device, and comprises a second acoustic signal generated by a second device, external to user device. (c) receive, from external source, a reference signal, indicative of a content included in the second signal. (d) in a case that a content of reference signal and the content in second signal are not time-synchronized, generate a synchronized reference signal, based on reference signal. Content of synchronized reference signal and second signal content are time-synchronized. (e) perform acoustic signal cancelling of that portion of captured signal which corresponds to second signal, based on reference signal. Generate reduced signal. This facilitates output of reduced signal to third device.

Abstract: A system and method for addressing distracting ambient noise during online collaborative meetings monitors when an online meeting is entered by a user at their computer. Smart devices are automatically powered down, paused or otherwise controlled during the online meeting. Devices are returned to their former state once the user leaves the meeting.

Abstract: An active noise reduction system includes a reference signal generator configured to generate a reference signal, a canceling sound generator configured to generate a canceling sound, an error detector configured to detect an error between a noise and the canceling sound and generate an error signal corresponding to the error, and a controller configured to control the canceling sound generator based on the reference signal and the error signal, wherein the controller is configured to update an estimation value of acoustic characteristics in an internal space of a mobile body based on the reference signal and the error signal, estimate a head position of an occupant in the internal space based on the updated estimation value of the acoustic characteristics, and update a control filter based on the estimated head position of the occupant, the control filter being a filter for controlling the canceling sound generator.

Abstract: Embodiments herein provide a system and a method for automatically generating at least one synthetic talking head video using a machine learning model. The method includes (i) extracting features from each frame of a video that is extracted from data sources, (ii) analyzing, using a face-detection model, the video to determine a driving face video if a number of identities, and faces of speakers are equal to one in all frames of the video, (iii) generating, using a text to speech model, synthetic speech utterances by automatically selecting a vocabulary of words and sentences from the data sources, (iv) modifying lip movements that are originally present in the driving face video corresponding to the synthetic speech utterances, and (v) generating, using machine learning model, synthetic talking head video based on the lip movements that are modified corresponding to the synthetic speech utterances.

Abstract: The present disclosure relates to a hearing apparatus comprising: a bone conduction sensor configured to convert bone vibrations of voice sound information into a bone conduction signal; a signal processing unit configured to implement a synthetic speech generation process, the synthetic speech generation process implementing a speech model; wherein the synthetic speech generation process receives the bone conduction signal as a control input and outputs a synthetic speech signal.

Abstract: An information processing apparatus includes a voice data input device (a microphone) to which voice data is input; a user data input device (an input device) to which user data is input from a user; and a controller. The controller recognizes first voice data input to the voice data input device (the microphone), outputs second voice data indicating that predetermined equipment is to be controlled based on a recognition result the first voice data, and controls the predetermined equipment based on the recognition result after the second voice data is notified, but stops controlling the predetermined equipment if user data indicating that control of the predetermined equipment is to be stopped is received during a time period when the second voice data is being notified.

Abstract: A method of detecting live speech comprises: receiving a signal containing speech; obtaining a first component of the received signal in a first frequency band, wherein the first frequency band includes audio frequencies; and obtaining a second component of the received signal in a second frequency band higher than the first frequency band. Then, modulation of the first component of the received signal is detected; modulation of the second component of the received signal is detected; and the modulation of the first component of the received signal and the modulation of the second component of the received signal are compared. It may then be determined that the speech may not be live speech, if the modulation of the first component of the received signal differs from the modulation of the second component of the received signal.

Abstract: Proposed are a speech recognition system and method for automatically calibrating a data label. A speech recognition method for automatically calibrating a data label according to an embodiment may comprise the steps of: performing confidence-based filtering to find the location of occurrence of a wrong label in time-series speech data, in which a correct label and the wrong label are temporally mixed, by using a transformer-based speech recognition model; and after performing filtering, replacing a label at a decoder time step, which has been determined to be a wrong label by the location of occurrence of the wrong label, so as to improve the performance of the transformer-based speech recognition model, wherein the step of performing confidence-based filtering to find the location of occurrence of the wrong label in the time-series speech data comprises finding and calibrating the wrong label using the confidence obtained by using a transition probability between labels at every decoder time step.

Abstract: A method for training a slot tagging model may, when an entity used for slot tagging is added, accurately perform slot tagging corresponding to the added entity only by adding new data to an external dictionary, without retraining, a computer-readable medium storing a program for performing the training method, a speech recognition apparatus providing a speech recognition service using the trained slot tagging model, and an electronic device used to provide the speech recognition service.

Abstract: Systems and methods for determining to perform an action of a query using a trained natural language model of a natural language understanding (NLU) system are disclosed herein. A text string corresponding to a prescribed action includes at least a content entity is received. A determination is made as to whether the text string corresponds to an audio input of a first group. In response to determining the text string corresponds to an audio input of a first group, a determination is made as to whether the text string includes an obsequious expression. In response to determining the text string corresponds to an audio input of a first group and in response to determining the text string includes an obsequious expression, a determination is made to perform the prescribed action.

Abstract: Methods, systems, and apparatus for receiving audio data corresponding to a user utterance and context data, identifying an initial set of one or more n-grams from the context data, generating an expanded set of one or more n-grams based on the initial set of n-grams, adjusting a language model based at least on the expanded set of n-grams, determining one or more speech recognition candidates for at least a portion of the user utterance using the adjusted language model, adjusting a score for a particular speech recognition candidate determined to be included in the expanded set of n-grams, determining a transcription of user utterance that includes at least one of the one or more speech recognition candidates, and providing the transcription of the user utterance for output.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for converting audio to spikes for input to a spiking neural network configured to recognize speech based on the spikes are described. In some aspects, a method includes obtaining audio data and generating frequency domain audio signals that represent the audio data by converting the audio data into a frequency domain. The frequency domain audio signals are mapped into a set of Mel-frequency bands to obtain Mel-scale frequency audio signals. A log transformation is performed on the Mel-scale frequency audio signals to obtain log-Mel signals. Spike input is generated for input to a spiking neural network (SNN) model by converting the log-Mel signals to the series of spikes. The spike input is provided as an input to the SNN model.

Abstract: An object of the present invention is to provide a system having a function of, in a case where there is ambiguity of being incapable of specifying an entity or content referred to by a noun or a noun equivalent expression in utterance exchanged in a communication system via a computer network, searching for accurate meaning or information serving as a clue to understand the meaning using only information regarding an utterance sentence or an utterer as a clue, instead of a response to a question sentence created by a user, and presenting a result to the user.

Abstract: A dialogue system includes a database, a voice recognition module configured to convert an utterance of a user in a vehicle into text, an intention determination module configured to identify intention of the user based on the text, an emotion determination module configured to identify the user's emotional state from the identified user's intention, and a controller configured to compare data indicating the identified intention of the user and the user's emotional state with rules stored in the database and determine whether to output a response to the user's utterance based on a result of the comparing.

Abstract: A control method for an electronic device is disclosed. The control method comprises the steps of during a call with a user of another electronic device by means of an electronic device, acquiring at least one keyword from a content of the call with the user of the another electronic device displaying the at least one keyword during the call and providing a search result for a keyword selected by a user from among the displayed at least one keyword. Particularly, at least a part of the control method of the present disclosure may use an artificial intelligence model learned according to at least one of machine learning, a neural network, and a deep learning algorithm.

Abstract: Methods and systems are presented for translating informal utterances into formal texts. Informal utterances may include words in abbreviation forms or typographical errors. The informal utterances may be processed by mapping each word in an utterance into a well-defined token. The mapping from the words to the tokens may be based on a context associated with the utterance derived by analyzing the utterance in a character-by-character basis. The token that is mapped for each word can be one of a vocabulary token that corresponds to a formal word in a pre-defined word corpus, an unknown token that corresponds to an unknown word, or a masked token. Formal text may then be generated based on the mapped tokens. Through the processing of informal utterances using the techniques disclosed herein, the informal utterances are both normalized and sanitized.

Abstract: An automatic speech recognition (ASR) system recognizes speech expressed in different languages. The ASR system includes a language-agnostic encoding component and prediction component. A language-specific joint analysis system generates first-language probabilities for symbols of a first language and second-language probabilities for symbols of a second language, based on outputs generated by the encoding component and the prediction component. The ASR system then modifies the probabilities produced by the joint analysis system by language-specific weighting information that, in turn, is produced by an attention system. This yields modified first-language probabilities and modified second-language probabilities. Finally, the ASR system predicts an updated instance of label information based on the modified first-language probabilities and the modified second-language probabilities. The ASR system can be successfully applied to recognize an utterance that combines words in two or more languages.

Abstract: Techniques for ensuring content output to a user conforms to a quality of the user's speech, even when a speechlet or skill ignores the speech's quality, are described. When a system receives speech, the system determines an indicator of the speech's quality (e.g., whispered, shouted, fast, slow, etc.) and persists the indicator in memory. When the system receives output content from a speechlet or skill, the system checks whether the output content is in conformity with the speech quality indicator. If the content conforms to the speech quality indicator, the system may cause the content to be output to the user without further manipulation. But, if the content does not conform to the speech quality indicator, the system may manipulate the content to render it in conformity with the speech quality indicator and output the manipulated content to the user.

Abstract: Technical solutions relate to the fields of artificial intelligence technologies and voice technologies. A technical solution includes: performing voice recognition and demand analysis on a voice instruction input by a user; in response to an unknown demand obtained by the demand analysis, acquiring information of a query entity and query content using a result of the demand analysis, and acquiring reply information corresponding to the query content by communication with the query entity; and returning a first voice response to the user using the reply information.

Abstract: A method includes: in response to identifying a primary user and corresponding Primary AI Assistant for a meeting, receiving by the Primary AI Assistant a confirmation to enroll at least one user personal digital assistant (PDA) of a respective one of at least one user; prompting the at least one user to provide descriptive information associated with the respective user PDA; connecting the at least one user PDA to the Primary AI Assistant internally by the Primary AI Assistant using the descriptive information for submitting requests; identifying by the Primary AI Assistant keywords and phrases received from the at least one user or primary user in the meeting; determining by the Primary AI Assistant a scheduling item based on the identified keywords and phrases; and automatically providing by the Primary AI Assistant the scheduling item to at least one user PDA corresponding to the scheduling item using the descriptive information.

Abstract: A voice controlled studio apparatus comprising a presenter interface unit and a producer interface unit, the presenter interface unit and the producer interface unit each adapted to generate commands and each unit comprising a voice input device, the apparatus further comprising a data processing unit adapted to receive commands from the presenter interface and the producer interface, process the commands, parse them to ascertain whether the actions meet at least one pre-determined criterion and then subsequently effect one or more actions based on the commands and the or each pre-determined criterion, and wherein the data processing unit is adapted to prioritise the effecting of actions so that commands generated by the producer interface unit can override the effecting of commands generated by the presenter interface unit, the apparatus further comprising a teleprompt unit adapted to provide a display adapted to be visible by a presenter and adapted to receive actions from a data processing unit and vary th

Abstract: Dynamic configuration and reconfiguration of Voice Control Assistants (VCAs) for Customer Premise Equipment (CPE) devices is provided via a Graphical User Interface (GUI). GUI is presented to provide a list of one or more voice control assistants (VCAs) supported by a CPE device. A selection of a VCA is received from the list of the one or more VCAs presented by the GUI. A configuration interface for the VCA selected from the list of the one or more VCAs from the list of the one or more VCAs is presented on the GUI. Configuration input on the configuration interface of the GUI is received to generate configuration data for the VCA selected from the list of the one or more VCAs. The configuration data is provided to the CPE for implementing the VCA selected from the list of the one or more VCAs according to the configuration data.

Abstract: Methods and systems for rendering lists of instructions and performing actions associated with those lists are described herein. In some embodiments, an individual may request that a voice activated electronic device associated with their user account assist in performing a task using a list of instructions. The list of instructions may include metadata that indicates actions capable of being performed by additional Internet of Things (“IoT”) devices. When the instructions are rendered, an instructions speechlet may recognize the metadata and may cause one or more of the IoT devices to perform a particular action. Furthermore, the metadata may also correspond to content capable of being rendered by the voice activated electronic device to assist the individual in performing a particular step of the instructions.

Abstract: An example method includes, at an electronic device: receiving an indication of a notification; in accordance with receiving the indication of the notification: obtaining one or more data streams from one or more sensors; determining, based on the one or more data streams, whether a user associated with the electronic device is speaking; and in accordance with a determination that the user is not speaking: causing an output associated with the notification to be provided.

Abstract: Various tools are disclosed for providing assistive or augmentative means to enhance the fluency and accuracy of persons having speech disabilities. These technologies may automatically ascertain and dynamically improve the accuracy with which automatic speech recognition (ASR) systems recognize utterances of persons having impaired speech conditions. In an embodiment, digitized audio information about a speaker’s utterance is processed to determine a set of candidate words matching the utterance. From these candidate words, a set of concepts is determined using a finite state machine model. A pictogram representing each concept is identified and presented to the speaker so that the speaker may select the pictogram corresponding to the best match of his or her intended meaning associated with the utterance. An action corresponding to speaker’s selection then may be performed. For example, displaying or synthesizing speech from textual information describing the selected concept.

Abstract: Provided here are systems and method for two-way authentication of a user. In embodiments, the method may include, in response to reception of an audio signal from a user, determining, via a smart device and/or authentication circuitry, whether a corresponding bone conduction signal is received from one of one or more separate wearable devices. The method may include, in response to determination that the corresponding bone conduction signal is received: determining, via the smart device and/or authentication circuitry, whether the audio signal and corresponding conduction signal are consistent. The method may include, in response to a determination that the audio signal and corresponding conduction signal are consistent: verifying, via the smart device, the audio signal; verifying, via the smart device, the corresponding bone conduction signal; and, in response to verification of the audio signal and the corresponding bone conduction signal, authenticating a user.

Abstract: Disclosed is a voice processing device. The voice processing device comprises a memory and a processor configured to perform sound source isolation on voice signals associated with the voices of speakers on the basis of the sound source positions of the respective voices. The processor is configured to: generate sound source position information indicating the sound source positions of the respective voices using the voice signals associated with the voices; generate isolated voice signals associated with the voices of the respective speakers from the voice signals on the basis of the sound source position information; and match the isolated voice signals and the voice source position information and store the same in the memory.

Abstract: Implementations generally relate to hearing aids. In some implementations, a method includes receiving sound at a hearing aid. The method further includes detecting a voice from the sound. The method further includes identifying the voice. The method further includes providing identity information associated with the voice.

Abstract: A system for generating channel-compensated features of a speech signal includes a channel noise simulator that degrades the speech signal, a feed forward convolutional neural network (CNN) that generates channel-compensated features of the degraded speech signal, and a loss function that computes a difference between the channel-compensated features and handcrafted features for the same raw speech signal. Each loss result may be used to update connection weights of the CNN until a predetermined threshold loss is satisfied, and the CNN may be used as a front-end for a deep neural network (DNN) for speaker recognition/verification. The DNN may include convolutional layers, a bottleneck features layer, multiple fully-connected layers, and an output layer. The bottleneck features may be used to update connection weights of the convolutional layers, and dropout may be applied to the convolutional layers.

Abstract: Implementations described herein relate to various techniques for biasing interpretations of spoken utterances that are received in a vehicular environment. For example, implementations can receive a spoken utterance that includes a query from a user of a vehicle and obtain a corresponding vehicle sensor data instance generated by vehicle sensor(s) of the vehicle. Some implementations can determine to execute a search over only a first corpus of data, but not a second corpus of data, to obtain a given response to the query based on various criteria, including at least the query, the corresponding vehicle sensor data instance, a corresponding timestamp associated with the corresponding vehicle sensor data instance, and/or a corresponding duration of time the user has been associated with the vehicle. Additional, or alternative, implementations can execute a search over both the first and second corpora of data, and obtain the given response based on the criteria.

Abstract: Implementations described herein relate to determining how to fulfill a spoken utterance based on a user that provided the spoken utterance. For example, implementations can receive a spoken utterance from a user, determine a set of fulfillment actions for the spoken utterance, and determine whether the user that provided the spoken utterance corresponds to a first user or a second user. Further, and in response to determining that the user corresponds to the first user, implementations can select a subset of first fulfillment action(s) from the set, and cause the subset of first fulfillment action(s) to be implemented to satisfy the spoken utterance. Moreover, and in response to determining that the user corresponds to the second user, implementations can select a subset of distinct, second fulfillment action(s) from the set, and cause the subset of second fulfillment action(s) to be implemented to satisfy the spoken utterance.

Abstract: An apparatus for improving context-based automatic interpretation performance includes: an uttered voice input unit configured to receive a voice signal from a user; a previous sentence input unit configured to determine whether there is a user’s previous utterance when the voice signal is input by the uttered voice input unit; a voice encoding processing unit configured to decode only the voice signal through the uttered voice input unit when it is determined that there is no user’s previous utterance and extract a vector of the voice signal when it is determined that there is the user’s previous utterance; a context encoding processing unit configured to extract a context vector from a previous utterance when there is the previous utterance and transmit the extracted context vector of the previous utterance; and an interpretation decoding processing unit configured to output an interpretation result text.

Abstract: Exemplary embodiments provide encoding and decoding methods, and associated encoders and decoders, for encoding and decoding of an audio scene which is represented by one or more audio signals. The encoder generates a bit stream which comprises downmix signals and side information which includes individual matrix elements of a reconstruction matrix which enables reconstruction of the one or more audio signals in the decoder.

Abstract: An audio encoder for encoding an audio signal having a lower frequency band and an upper frequency band includes: a detector for detecting a peak spectral region in the upper frequency band of the audio signal; a shaper for shaping the lower frequency band using shaping information for the lower band and for shaping the upper frequency band using at least a portion of the shaping information for the lower band, wherein the shaper is configured to additionally attenuate spectral values in the detected peak spectral region in the upper frequency band; and a quantizer and coder stage for quantizing a shaped lower frequency band and a shaped upper frequency band and for entropy coding quantized spectral values from the shaped lower frequency band and the shaped upper frequency band.

Abstract: A method for dynamically controlling enhancement of an audio stream is provided, where the audio stream defines a sequence of audio segments over time. Each audio segment defines a waveform having a plurality of waveform attributes. For each audio segment of the sequence of audio segments, the method includes: (i) determining a set of waveform-attribute values of the audio segment's waveform attributes, (ii) computing a first distance between the determined set of waveform-attribute values and a first predefined set of waveform-attribute values representative of speech, and computing a second distance between the determined set of waveform-attribute values and a second predefined set of waveform-attribute values representative of music, (iii) using the computed first and second distances as a basis to classify the audio segment as primarily speech or rather primarily music, and (iv) controlling, based on the classifying, whether or not to enhance the audio segment for output.

Abstract: Described are methods of processing audio data for hum noise detection and/or removal. The audio data comprises a plurality of frames. One method incudes: classifying frames of the audio data as either content frames or noise frames, using one or more content activity detectors; determining a noise spectrum from one or more frames of the audio data that are classified as noise frames; determining one or more hum noise frequencies based on the determined noise spectrum; generating an estimated hum noise signal based on the one or more hum noise frequencies; and removing hum noise from at least one frame of the audio data based on the estimated hum noise signal. Also described are apparatus for carrying out the methods, as well as corresponding programs and computer-readable storage media.