Abstract: Operations are appropriately changed in accordance with the use method. A keyword detection unit 11 generates a keyword detection result indicating a result of detecting an utterance of a predetermined keyword from an input voice. A voice detection unit 12 generates a voice section detection result indicating a result of detecting a voice section from the input voice. A sequential utterance detection unit 13 generates a sequential utterance detection result indicating that a sequential utterance has been made if the keyword detection result indicates that the keyword has been detected and if the voice section detection result indicates that the voice section has been detected.

Abstract: Systems and methods provide input and output mode control for audio processing on a user device. Audio processing may be configured by monitoring audio activity on a device having at least one microphone and a digital audio processing unit, collecting information from the monitoring of the activity, including an identification of at least one application utilizing audio processing, and determining a context for the audio processing, the context including at least one of a hardware, software, audio signal and/or environmental context. An audio signal processing configuration is determined based on the application and determined context, an associated audio signal processing mode is selected, and an optimized audio signal generated.

Abstract: An electronic device is provided. The electronic device according to an embodiment includes a microphone, a communicator comprising communication circuitry, and a processor configured to control the communicator to transmit a control command to an external audio device for reducing an audio output level of the external audio device in response to a trigger signal for starting a voice control mode being received through the microphone and to control the electronic device to operate in the voice control mode.

Abstract: Method includes receiving, through a plurality of channels, audio data corresponding to a plurality of frequency ranges; determining, for each channel's frequency ranges, speech audio and/or noise energy level using a model trained by machine learning; determining a speech signal with removed noise for each channel; determining one or more statistical values associated with an energy level of a channel's speech signal with the removed noise; determining a strongest channel that has highest statistical values associated with an energy level of a speech signal; determining that the one or more statistical values associated with the energy level of the strongest channel's speech signal satisfy a threshold condition; comparing statistical values associated with an energy level of a speech signal of each channel with those of the strongest channel; and determining whether to update a gain value for a channel based on the channel's statistical values associated with the energy level.

Abstract: A reminder method and apparatus and an electronic device. When a difference between a voice volume of a user in conversation and a reference volume is greater than a preset threshold, output of reminder to the user is controlled. Thus, the user may be reminded when his/her conversation is too loud for the surrounding environment automatically according to a comparing result of the voice volume of the user in conversation and the reference volume, and user experience may be improved.

Abstract: A method for compensating hearing deficiencies with a hearing device includes receiving a sound signal; attenuating an output sound pressure level of the sound signal dependent on an input sound pressure level of the sound signal; and outputting the attenuated sound signal with a loudspeaker of the hearing device; wherein the output sound pressure level is attenuated, when the input sound pressure level is above an upper speech recognition kneepoint of a user, which upper speech recognition kneepoint is stored in the hearing device and which has been selected dependent on a sound pressure level dependent speech recognition ability of the user.

Abstract: A method for speech level adaptation, the method includes: (A) Receiving, by a voice interactive intelligent personal assistant, multiple input audio signals that includes first and second groups of input audio signals. (B) Buffering the multiple input audio signals. (C) Searching for a voice trigger in the first group of input audio signals. When finding the voice trigger then (D) Determining a linear gain factor to be applied on the second group of buffered audio signals, (E) Applying the linear gain factor on the second group of buffered audio signals to provide the output audio signals; and (F) applying a speech recognition process on the output audio signals to detect an audio command embedded in the output audio signals.

Abstract: The present application discloses systems and methods to present information to recreational vehicle riders and to provide customizable visual information to recreational vehicle riders. The present application further discloses systems and methods to connect and transmit audio information between a driver portable communication device and a driver audio interface device through the recreational vehicle and audio information between a passenger portable communication device and a passenger audio interface device.

Abstract: A system for recording vehicle occupants and their immediate environment includes a driver-facing camera for capturing video data, an input that receives data from one or more sensors configured to detect vehicle related data and a processor. The processor is configured to embed, based on the sensor data, at least one indicator in a portion of the video data. The indicator designates that the portion of the video data is to be obscured during playback of the video data.

Abstract: A method for encoding an audio signal, comprising using one or more algorithms operating on a processor to filter the audio signal into two output signals, wherein each output signal has a sampling rate that is equal to a sampling rate of the audio signal, and wherein one of the output signals includes high frequency data. Using one or more algorithms operating on the processor to window the high frequency data by selecting a set of the high frequency data. Using one or more algorithms operating on the processor to determine a set of linear predictive coding (LPC) coefficients for the windowed data. Using one or more algorithms operating on the processor to generate energy scale values for the windowed data. Using one or more algorithms operating on the processor to generate an encoded high frequency bitstream.

Abstract: A method and an apparatus for synthesizing an audio signal are described. A spectral tilt is applied to the code of a codebook used for synthesizing a current frame of the audio signal. The spectral tilt is based on the spectral tilt of the current frame of the audio signal. Further, an audio decoder operating in accordance with the inventive approach is described.

Abstract: The present application discloses systems and methods to present information to recreational vehicle riders and to provide customizable visual information to recreational vehicle riders. The present application further discloses systems and methods to connect and transmit audio information between a driver portable communication device and a driver audio interface device through the recreational vehicle and audio information between a passenger portable communication device and a passenger audio interface device.

Abstract: A synchronised soundtrack for an audiobook. The soundtrack has a soundtrack timeline having one or more audio regions that are configured for synchronised playback with corresponding narration regions in the audiobook playback timeline. Each audio region having a position along the soundtrack timeline that is dynamically adjustable to maintain synchronization of the audio regions of the soundtrack with their respective narration regions in the audiobook based on a narration speed variable indicative of the playback narration speed of the audiobook.

Abstract: Methods, apparatus, systems and articles of manufacture are disclosed to adjust audio playback settings. Example apparatus disclosed herein include an equalization (EQ) model query generator to generate a query to a neural network, the query including a representation of a sample of an audio signal; an EQ filter settings analyzer to: access a plurality of audio playback settings determined by the neural network based on the query; and determine a filter coefficient to apply to the audio signal based on the plurality of audio playback settings; an EQ personalization manager to; generate a personalized EQ setting; and an EQ adjustment implementor to: blend the personalized EQ setting and the filter coefficient to generate a blended equalization; and apply the blended equalization to the audio signal in a first duration.

Abstract: Apparatus for decoding an encoded audio signal including an encoded core signal and parametric data, including: a core decoder for decoding the encoded core signal to obtain a decoded core signal; an analyzer for analyzing the decoded core signal before or after performing a frequency regeneration operation to provide an analysis result; and a frequency regenerator for regenerating spectral portions not included in the decoded core signal using a spectral portion of the decoded core signal, the parametric data, and the analysis result.

Abstract: Methods, apparatus, systems and articles of manufacture are disclosed to adjust audio playback settings based on analysis of audio characteristics. Example apparatus disclosed herein include an equalization (EQ) model query generator to generate a query to a neural network, the query including a representation of a sample of an audio signal; an EQ filter settings analyzer to: access a plurality of audio playback settings determined by the neural network based on the query; and determine a filter coefficient to apply to the audio signal based on the plurality of audio playback settings; and an EQ adjustment implementer to apply the filter coefficient to the audio signal in a first duration.

Abstract: An audio signal transmission device for encoding an audio signal includes an audio encoding unit that encodes an audio signal and a side information encoding unit that calculates and encodes side information from a look-ahead signal. An audio signal receiving device for decoding an audio code and outputting an audio signal includes: an audio code buffer that detects packet loss based on a received state of an audio packet, an audio parameter decoding unit that decodes an audio code when an audio packet is correctly received, a side information decoding unit that decodes a side information code when an audio packet is correctly received, a side information accumulation unit that accumulates side information obtained by decoding a side information code, an audio parameter missing processing unit that outputs an audio parameter upon detection of audio packet loss, and an audio synthesis unit that synthesizes decoded audio from the audio parameter.

Abstract: An apparatus for decomposing an audio signal into a background component signal and a foreground component signal includes: a block generator for generating a time sequence of blocks of audio signal values; an audio signal analyzer for determining a block characteristic of a current block of the audio signal and for determining an average characteristic for a group of blocks, the group of blocks including at least two blocks; and a separator for separating the current block into a background portion and a foreground portion in response to a ratio of the block characteristic of the current block and the average characteristic of the group of blocks, wherein the background component signal includes the background portion of the current block and the foreground component signal includes the foreground portion of the current block.

Abstract: Sensors, sensor controllers, and sensor control methods may employ an echo-magnification technique to improve threshold-based echo detection. In one illustrative embodiment, a sensor controller includes: a transmitter, a receiver, and a processing circuit coupled to the transmitter and to the receiver. The transmitter drives a piezoelectric element to generate acoustic bursts. The receiver senses a response of the piezoelectric element to echoes of each acoustic burst. The processing circuit is operable to apply echo-detection processing to the response by: identifying an interval of the response representing at least a portion of a potential echo; deriving a modified response from the response by selectively magnifying the response during said interval; and using the modified response to detect an echo.

Abstract: An audio signal transmission device for encoding an audio signal includes an audio encoding unit that encodes an audio signal and a side information encoding unit that calculates and encodes side information from a look-ahead signal. An audio signal receiving device for decoding an audio code and outputting an audio signal includes: an audio code buffer that detects packet loss based on a received state of an audio packet, an audio parameter decoding unit that decodes an audio code when an audio packet is correctly received, a side information decoding unit that decodes a side information code when an audio packet is correctly received, a side information accumulation unit that accumulates side information obtained by decoding a side information code, an audio parameter missing processing unit that outputs an audio parameter upon detection of audio packet loss, and an audio synthesis unit that synthesizes decoded audio from the audio parameter.

Abstract: A method may include obtaining a first audio signal including first speech originating at a remote device during a communication session between the remote device and a communication device and obtaining a second audio signal including second speech originating at the communication device during the communication session between the remote device and the communication device. The method may also include obtaining a characteristic of the communication session from one or more of: the first audio signal, the second audio signal, and settings of the communication device and determining a hearing level of a user of the communication device using the characteristic of the communication session.

Abstract: In an example implementation according to aspects of the present disclosure, a method may include identifying, via a first microphone of a device, when a user registered to use the device is speaking, and comparing a voice characteristic of the user, as detected by the first microphone, against a threshold value. If the voice characteristic exceeds the threshold value, the method may include unmuting a second microphone of the device for the user to participate in a teleconference.

Abstract: A robot apparatus including an input unit to receive a voice command from a user, a determination unit to determine whether a voice command is repeated a predetermined number of times, and a control unit to register a shortcut command to shorten a voice command if it is determined a voice command is repeated a predetermined number of times. A shortcut command to shorten a voice command of a user is generated, and thus user convenience is enhanced.

Abstract: Coordinating signal processing among computing devices in a voice-driven computing environment is provided. A first and second digital assistant can detect an input audio signal, perform a signal quality check, and provide indications that the first and second digital assistants are operational to process the input audio signal. A system can select the first digital assistant for further processing. The system can receive, from the first digital assistant, data packets including a command. The system can generate, for a network connected device selected from a plurality of network connected devices, an action data structure based on the data packets, and transmit the action data structure to the selected network connected device.

Abstract: Provided are methods and systems of using division-free duplexing (DFD) in a cable communication network. Techniques for applying DFD in a cable communication network may enable data to be transmitted and received over a coaxial cable without using division duplexing techniques. For example, the cable communication network may include DFD enabled network nodes and each subscriber to the cable network may be equipped with a DFD system configured to operate in a DFD mode. In some embodiments, oppositely propagating signals may be transmitted over one frequency channel, and DFD techniques may be used to recover originally transmitted signals. Further, in some embodiments, DFD techniques may be used with encryption methods to increase the security of data transmitted in the cable communication network.

Abstract: An apparatus and method for broadcast signal frame using layered division multiplexing are disclosed. An apparatus for generating broadcast signal frame according to an embodiment of the present invention includes a combiner configured to generate a multiplexed signal by combining a core layer signal and an enhanced layer signal at different power levels; a power normalizer configured to reduce the power of the multiplexed signal to a power level corresponding to the core layer signal; a time interleaver configured to generate a time-interleaved signal by performing interleaving that is applied to both the core layer signal and the enhanced layer signal; and a frame builder configured to generate a broadcast signal frame including a preamble for signaling time interleaver information shared by the core layer signal and the enhanced layer signal, using the time-interleaved signal.

Abstract: A system and method for transmitting a plurality of signals. The system includes a transmitting terminal configured to obtain and pack the plurality of signals in multiple signal sequences, with each of the plurality of signals having respective predefined periods. A receiving terminal is configured to receive and unpack the multiple signal sequences from the transmitting terminal. A controller is in communication with the transmitting terminal and the receiving terminal and configured to store a look-up table. The look-up table includes a respective pre-identifier and respective identifier for each of the plurality of signals. The controller is configured to determine if the respective predefined periods vary in magnitude within a frame. If so, the controller is configured to obtain the respective pre-identifiers and respective identifiers for each of the plurality of signals. The multiple signal sequences are based at least partially on the respective pre-identifiers and the respective identifiers.

Abstract: A method or associated system for motion adaptive speech processing includes dynamically estimating a motion profile that is representative of a user's motion based on data from one or more resources, such as sensors and non-speech resources, associated with the user. The method includes effecting processing of a speech signal received from the user, for example, while the user is in motion, the processing taking into account the estimated motion profile to produce an interpretation of the speech signal. Dynamically estimating the motion profile can include computing a motion weight vector using the data from the one or more resources associated with the user, and can further include interpolating a plurality of models using the motion weight vector to generate a motion adaptive model. The motion adaptive model can be used to enhance voice destination entry for the user and re-used for other users who do not provide motion profiles.

Abstract: A signal decoding method and device, where the method includes decoding a bit stream of a voice signal or an audio signal to acquire a decoded signal, predicting an excitation signal of an extension band according to the decoded signal, where the extension band is adjacent to a band of the decoded signal, and the band of the decoded signal is lower than the extension band; selecting a first band and a second band from the decoded signal, and predicting a spectral envelope of the extension band according to a spectral coefficient of the first band and a spectral coefficient of the second band; and determining a frequency-domain signal of the extension band according to the spectral envelope of the extension band and the excitation signal of the extension band.

Abstract: A robot apparatus including an input unit to receive a voice command from a user, a determination unit to determine whether a voice command is repeated a predetermined number of times, and a control unit to register a shortcut command to shorten a voice command if it is determined a voice command is repeated a predetermined number of times. A shortcut command to shorten a voice command of a user is generated, and thus user convenience is enhanced.

Abstract: There is provided a state-monitory system of a moving apparatus which is capable of deducing an abnormality and cause of breakdown of the moving apparatus by Equationematical analysis only by measuring vibration of a casing or object to be monitored and which requires no large-scaled apparatus. An equivalent model of the moving apparatus 1 is formed, which is adapted to replace the vibration of a monitored object of the moving apparatus 1 with the vibration of a viscoelastic member such as an elastic spring etc. and which is adapted to be capable of outputting virtual vibration data having synthesized vibration equivalent to the vibration of the viscoelastic member, and the equivalent model is Equationematically analyzed, thereby estimating the abnormality and cause of breakdown of the moving apparatus 1 only by measuring vibration of the casing 8 or a monitored object.

Abstract: A process for controlling a plurality of mobile-radio equipped robots in a talkgroup includes receiving, at a mobile-radio equipped robot via a wireless communications interface comprising one of an infrastructure wireless communication interface for communicating with an infrastructure radio access network (RAN) and an ad-hoc wireless communication interface for communicating with an ad-hoc network, a group voice call containing a voice-command. The mobile-radio equipped robot determines that it is a target of the group voice call, and responsively text-converts the voice-command into an actionable text-based command. The mobile-radio equipped robot subsequently operates a mechanical drive element in accordance with the actionable text-based command.

Abstract: The invention relates to an audio signal processing apparatus for processing an input earpiece audio signal upon the basis of a microphone audio signal, the audio signal processing apparatus comprising a voice activity detector being configured to determine a voice activity indicator signal upon the basis of the input earpiece audio signal, a noise magnitude determiner being configured to determine a microphone noise magnitude indicator signal upon the basis of the microphone audio signal, a gain factor determiner being configured to determine a gain factor signal upon the basis of the voice activity indicator signal and the microphone noise magnitude indicator signal, and a weighter being configured to weight the input earpiece audio signal by the gain factor signal to obtain an output earpiece audio signal.

Abstract: A gain control system for controlling gain applied to an audio signal includes a power estimator configured to estimate the power of a digital signal derived from the audio signal, a digital gain estimator configured to determine, in dependence on the estimated power, a digital gain which would modify the power of the digital signal so as to reach a target power level, and a gain controller configured to adjust an analog gain applied to the audio signal in dependence on the determined digital gain.

Abstract: An electronic circuit includes a peak detector, a gain controller, and a compressor. The peak detector detects a peak level from a digital input. The gain controller outputs a digital gain with regard to increasing the peak level to a target level, The compressor provides the gain controller with a compressed gain which is to be output as the digital gain, based on the detected peak level. In a compression interval where the peak level is greater than a threshold level, the output digital gain increases as the peak level decreases. The compressor generates the compressed gain such that a ratio of an increment of the output digital gain to a decrement of the peak level in the compression interval is less than a reference ratio.

Abstract: Embodiments of the present invention provide a voiceprint verification method and device. The voiceprint verification method comprises receiving verification voice data relating to a verification phrase; generating a verification voiceprint on the basis of said verification voice data; determining whether a similarity value between the verification voiceprint and a reference voiceprint conforms to a preset similarity value; and, if the similarity value between the verification voiceprint and a reference voiceprint conforms to a preset similarity value, then determining there is a match between the verification voiceprint and the reference voiceprint.

Abstract: Speech in a motor vehicle is improved by suppressing transient, “non-stationary” noise using pattern matching. Pre-stored sets of linear predictive coefficients are compared to LPC coefficients of a noise signal. The pre-stored LPC coefficient set that is “closest” to an LPC coefficient set representing a signal comprising speech and noise is considered to be noise.

Abstract: An audio coding device includes a filter configured to extract a low-band signal having a first frequency component from an input signal, a memory, and a processor coupled to the memory and configured to extract envelope information relating to an envelope of a high-band signal having a second frequency component which is higher than the first frequency component in the input signal, detect tone information that is information on a tone signal included in a high-band signal spectrum from the input signal, correct the envelope information based on a difference between frequency of the tone signal and frequency of a peak of the envelope, and code the low-band signal, the tone information, and the envelope information that is corrected.

Abstract: A process for controlling a plurality of mobile-radio equipped robots in a talkgroup includes receiving, at a mobile-radio equipped robot via a wireless communications interface comprising one of an infrastructure wireless communication interface for communicating with an infrastructure radio access network (RAN) and an ad-hoc wireless communication interface for communicating with an ad-hoc network, a group voice call containing a voice-command. The mobile-radio equipped robot determines that it is a target of the group voice call, and responsively text-converts the voice-command into an actionable text-based command. The mobile-radio equipped robot subsequently operates a mechanical drive element in accordance with the actionable text-based command.

Abstract: An improved gain-shape vector quantization is achieved by determining a number of bits to be allocated to a gain adjustment- and shape-quantizer for a plurality of combinations of a current bit rate and a first signal property. The bit allocation is derived by using an average of optimal bit allocations for a training data set. A number of bits to the gain adjustment and the shape quantizers for a plurality of combinations of the bit rate and a first signal are pre-calculated, and a table indicating the number of bits to be allocated to the gain adjustment- and the shape-quantizers for a plurality of combinations of the bit rate and a first signal property is created. In this way, the table can be used for achieving an improved bit allocation.

Abstract: A method includes identifying, by a computing device including a processor, a plurality of words from data. The method includes determining a location of the computing device. The method includes determining, by the computing device, a sound output level based on the location. The method also includes generating, by the computing device, digital sound data based on the sound output level and the plurality of words identified from the data.

Abstract: Techniques for indicating to a voice-controlled device that a user is going to provide a voice command to the device. In response to receiving such an indication, the device may prepare to process an audio signal based on sound captured by a microphone of the device for the purpose of identifying the voice command from the audio signal. For instance, a user may utilize a signaling device that includes a button that, when actuated, sends a signal that is received by the voice-controlled device. In response to receiving the signal, a microphone of the voice-controlled device may capture sound that is proximate to the voice-controlled device and may create an audio signal based on the sound. The voice-controlled device may then analyze the audio signal for a voice command of the user or may provide the audio signal to a remote service for identifying the command.

Abstract: Personal audio systems and methods are disclosed. A personal audio system includes a voice activity detector to determine whether or not an ambient audio stream contains voice activity, a pitch estimator to determine a frequency of a fundamental component of an annoyance noise contained in the ambient audio stream, and a filter bank to attenuate the fundamental component and at least one harmonic component of the annoyance noise to generate a personal audio stream. The filter bank implements a first filter function when the ambient audio stream does not contain voice activity, or a second filter function when the ambient audio stream contains voice activity.

Abstract: A sound emission and collection device includes a speaker, a filter processing a sound emission signal, microphones, echo cancellers cancelling regression sound signals of the sound emitted by the speaker from the sound collection signals of the corresponding microphones, a first integration section integrating adaptive filter coefficients taken out from the plurality of echo cancellers, a reverberation time estimation section estimating the reverberation time for each frequency band in the space in which the speaker and the plurality of microphones are present on the basis of the integrated adaptive filter coefficient, and an arithmetic operation section specifying a frequency band having a long reverberation time from the sound emission signal based on the estimated reverberation time, calculating a filter coefficient for suppressing power of the specified frequency band, and setting the filter coefficient to the filter.

Abstract: Vision-assist devices and methods are disclosed. In one embodiment, a vision-assist device includes an image sensor for generating image data corresponding to a scene, a processor, and an auditory device. The processor is programmed to receive the image data from the image sensor, perform object recognition on the image data to determine a classification of a detected object that is present within the scene, and determine a confidence value with respect to the classification of the detected object. The confidence value is based on a confidence that the classification of the detected object matches an actual classification of the detected object. The processor further generates an auditory signal based on the confidence value. The audio device receives the auditory signal from the processor and produces an auditory message from the auditory signal. The auditory message is indicative of the classification of the detected object and the confidence value.

Abstract: Embodiments include methods and systems for improving an acoustic model. Aspects include acquiring a first standard deviation value by calculating standard deviation of a feature from first training data and acquiring a second standard deviation value by calculating standard deviation of a feature from second training data acquired in a different environment from an environment of the first training data. Aspects also include creating a feature adapted to an environment where the first training data is recorded, by multiplying the feature acquired from the second training data by a ratio obtained by dividing the first standard deviation value by the second standard deviation value. Aspects further include reconstructing an acoustic model constructed using training data acquired in the same environment as the environment of the first training data using the feature adapted to the environment where the first training data is recorded.

Abstract: Embodiments include methods and systems for improving an acoustic model. Aspects include acquiring a first standard deviation value by calculating standard deviation of a feature from first training data and acquiring a second standard deviation value by calculating standard deviation of a feature from second training data acquired in a different environment from an environment of the first training data. Aspects also include creating a feature adapted to an environment where the first training data is recorded, by multiplying the feature acquired from the second training data by a ratio obtained by dividing the first standard deviation value by the second standard deviation value. Aspects further include reconstructing an acoustic model constructed using training data acquired in the same environment as the environment of the first training data using the feature adapted to the environment where the first training data is recorded.

Abstract: An audio peak limiter apparatus which calculates a smoothed sequence of gains for application to a sequence of blocks of samples of an audio signal. The apparatus sometimes calculates a candidate gain, to replace a too-large smoothed gain, such that applying the candidate gain would produce no scaled sample whose magnitude exceeds a predetermined limit. The apparatus sometimes calculates and stores a final gain to replace the too-large smoothed gain (where the final gain can be obtained e.g. by dividing the to-be-replaced smoothed gain by a prior smoothed gain and multiplying the result by the corresponding prior final gain), if it is determined that the candidate gain is not less than the immediately previous final gain.

Abstract: A method of performing gain adjustment in an electronic device includes determining a first set of spectral frequency values and determining a second set of spectral frequency values. The first set of spectral frequency values corresponds to a high-band portion of an audio signal received at the electronic device. The second set of spectral frequency values approximates the first set of spectral frequency values in the high band portion of the audio signal. The method includes estimating a spectral distortion corresponding to a difference between the first set of spectral frequency values and the second set of spectral frequency values and adjusting, based on the spectral distortion, a gain value corresponding to at least a portion of the audio signal. The method also includes transmitting an encoded bitstream that includes information corresponding to the adjusted gain value and the second set of spectral frequency values.

Abstract: Embodiments include methods and systems for improving an acoustic model. Aspects include acquiring a first standard deviation value by calculating standard deviation of a feature from first training data and acquiring a second standard deviation value by calculating standard deviation of a feature from second training data acquired in a different environment from an environment of the first training data. Aspects also include creating a feature adapted to an environment where the first training data is recorded, by multiplying the feature acquired from the second training data by a ratio obtained by dividing the first standard deviation value by the second standard deviation value. Aspects further include reconstructing an acoustic model constructed using training data acquired in the same environment as the environment of the first training data using the feature adapted to the environment where the first training data is recorded.