Abstract: A method for processing a digital signal, implemented during decoding of the signal, in order to replace a succession of samples lost during decoding, the method comprising steps of: generating a structure of a signal for replacing the lost succession, this structure comprising spectral components determined from valid samples received during decoding before the succession of lost samples; generating a residue between a digital signal available to the decoder, comprising received valid samples, and a signal generated from the spectral components; and extracting blocks from the residue, method in which window weighted blocks are injected into the structure using an overlap-add approach, the injected blocks partially overlapping in time.

Abstract: An apparatus and method are disclosed for filtering an audio signal are disclosed. The apparatus includes an analysis filter bank, a high frequency reconstructor or a parametric stereo processor, and a synthesis filter bank. The analysis filterbank receives real-valued time domain input audio samples and generates complex valued subband samples. The high frequency reconstructor or parametric stereo processor modifies at least some of the complex valued subband samples. The synthesis filter bank receives the modified complex valued subband samples and generates time domain output audio samples. The analysis filter bank comprises analysis filters that are complex exponential modulated versions of a prototype filter.

Abstract: Some embodiments relate to techniques for encoding an audio signal represented by a plurality of frames including a first frame. The techniques include using at least one computer hardware processor to perform: obtaining an initial discrete spectral representation of the first frame; obtaining a primary discrete spectral representation of the initial discrete spectral representation at least in part by estimating a phase envelope of the initial discrete spectral representation and evaluating the estimated phase envelope at a discrete set of frequencies; calculating a residual discrete spectral representation of the initial discrete spectral representation based on the initial discrete spectral representation and the primary discrete spectral representation; and encoding the residual discrete spectral representation using a plurality of codewords.

Abstract: Systems, methods, and computer-readable storage devices for crowd-sourced data labeling. The system requests a respective response from each of a set of entities. The set of entities includes crowd workers. Next, the system incrementally receives a number of responses from the set of entities until one of an accuracy threshold is reached and m responses are received, wherein the accuracy threshold is based on characteristics of the number of responses. Finally, the system generates an output response based on the number of responses.

Abstract: From a plurality of received voice signals, a signal interval in which there is a talker collision between at least a first and a second voice signal is detected. A processor receives a positive detection result and processes, in response to this, at least one of the voice signals with the aim of making it perceptually distinguishable. A mixer mixes the voice signals to supply an output signal, wherein the processed signal(s) replaces the corresponding received signals. In example embodiments, signal content is shifted away from the talker collision in frequency or in time. The invention may be useful in a conferencing system.

Abstract: Disclosed herein are systems, methods, and tangible computer readable-media for detecting synthetic speaker verification. The method comprises receiving a plurality of speech samples of the same word or phrase for verification, comparing each of the plurality of speech samples to each other, denying verification if the plurality of speech samples demonstrate little variance over time or are the same, and verifying the plurality of speech samples if the plurality of speech samples demonstrates sufficient variance over time. One embodiment further adds that each of the plurality of speech samples is collected at different times or in different contexts. In other embodiments, variance is based on a pre-determined threshold or the threshold for variance is adjusted based on a need for authentication certainty. In another embodiment, if the initial comparison is inconclusive, additional speech samples are received.

Abstract: An audio processing device comprises a multitude of electric input signals, each electric input signal being provided in a digitized form, and a control unit receiving said digitized electric input signals and providing a resulting enhanced signal. The control unit is configured to determine the resulting enhanced signal from said digitized electric input signals, or signals derived therefrom, according to a predefined scheme.

Abstract: Embodiments of the present invention provide a stereo signal down-mixing method, encoding/decoding apparatus and system. The down-mixing method includes: converting a first channel time-domain signal and a second channel time-domain signal into a first channel frequency-domain signal and a second channel frequency-domain signal; obtaining a frequency-domain channel signal level difference and a frequency-domain channel signal phase difference between the two channel frequency-domain signals; for each frequency bin in each frequency band, using a function based on the frequency-domain channel signal level difference and frequency-domain channel signal phase difference to obtain a down-mixed signal phase that is located between phases of the two channel frequency-domain signals, and obtaining a down-mixed signal amplitude through calculation; and obtaining a frequency-domain down-mixed signal according to the phase and amplitude.

Abstract: A voice coding device capable of preventing overall quality degradation even when the bit rate for coding is lowered. The voice coding device codes a wide band signal in a first layer, and codes an extended band signal whose frequency band is located in higher frequency than the wide band signal in an extended band layer. An adaptive band selection unit (301) selects a frequency band to be excluded from a coding object in the extended band layer or a frequency band whose energy is to be attenuated in the extended band layer. A band-limited signal generation unit (302) excludes, within the frequency band of an input signal, the frequency band selected by the adaptive band selection unit (301) from the coding object, or attenuates the energy of the frequency band selected by the adaptive band selection unit (301).

Abstract: An apparatus determining a weighting function for line prediction coding coefficients quantization converts a linear prediction coding (LPC) coefficient of an input signal into one of a line spectral frequency (LSF) coefficient and an immitance spectral frequency (ISF) coefficient and determines a weighting function associated with one of an importance of the ISF coefficient and importance of the LSF coefficient using one of the converted ISF coefficient and the converted LSF coefficient.

Abstract: A method and apparatus for sharing information in a video decoding system are disclosed. The method derives reconstructed data for a picture from a bitstream, where the picture is partitioned into multiple slices. An information-sharing flag is parsed from the bitstream associated with a current reconstructed slice. If the information-sharing flag indicates information sharing, shared information is determined from a part of the bitstream not corresponding to the current reconstructed slice, and in-loop filtering process is applied to the current reconstructed slice according to the shared information. If the information-sharing flag indicates filter no information sharing, individual information is determined from a part of the bitstream corresponding to the current reconstructed slice, and in-loop filtering process is applied to the current reconstructed slice according to the individual information. A method for a corresponding encoder is also disclosed.

Abstract: Beamforming precoding matrix using non-uniform angles quantization. Adaptively generated feedback information is provided between communication devices that communicate using more than one communication path, link, connection, etc. With respect to feeding back different types of information having different respective characteristics (e.g., different respective probability density functions), different and respective quantization may be employed for the different types of information. For example, uniform, Gaussian, or per bit loop optimized quantization may be individually selected and employed for each of the different types of feedback information used in a wired communication system (e.g.

Abstract: The present invention is based on the finding that parameters including: a first set of parameters of a representation of a first portion of an original signal and a second set of parameters of a representation of a second portion of the original signal can be efficiently encoded when the parameters are arranged in a first sequence of tuples and a second sequence of tuples. The first sequence of tuples includes tuples of parameters having two parameters from a single portion of the original signal and the second sequence of tuples includes tuples of parameters having one parameter from the first portion and one parameter from the second portion of the original signal. A bit estimator estimates the number of necessary bits to encode the first and the second sequence of tuples. Only the sequence of tuples, which results in the lower number of bits, is encoded.

Abstract: In a method of improving perceived loudness and sharpness of a reconstructed speech signal delimited by a predetermined bandwidth, performing the steps of providing (S10) the speech signal, and separating (S20) the provided signal into at least a first and a second signal portion. Subsequently, adapting (S30) the first signal portion to emphasize at least a predetermined frequency or frequency interval within the first bandwidth portion. Finally, reconstructing (S40) the second signal portion based on at least the first signal portion, and combining (S50) the adapted first signal portion and the reconstructed second signal portion to provide a reconstructed speech signal with an overall improved perceived loudness and sharpness.

Abstract: The invention provides a method for estimating a fundamental frequency of a speech signal comprising the steps of receiving a signal spectrum of the speech signal, filtering the signal spectrum to obtain a refined signal spectrum, determining a cross-power spectral density using the refined signal spectrum and the signal spectrum, transforming the cross-power spectral density into the time domain to obtain a cross-correlation function, and estimating the fundamental frequency of the speech signal based on the cross-correlation function.

Abstract: MDCT or FFT-based audio coding algorithms often have the problem named here spectral pre-echoes when coding an energy attack signal. This invention presents several possibilities to avoid the spectral pre-echoes existing in decoded signal segment before the energy attack point. The spectral envelope before the attack point can be improved by performing spectrum smoothing, replacing the segment of having spectral pre-echoes or filtering the segment with a combined filter obtained by doing LPC analysis.

Abstract: A mixed time-domain/frequency-domain coding device and method for coding an input sound signal, wherein a time-domain excitation contribution is calculated in response to the input sound signal. A cut-off frequency for the time-domain excitation contribution is also calculated in response to the input sound signal, and a frequency extent of the time-domain excitation contribution is adjusted in relation to this cut-off frequency. Following calculation of a frequency-domain excitation contribution in response to the input sound signal, the adjusted time-domain excitation contribution and the frequency-domain excitation contribution are added to form a mixed time-domain/frequency-domain excitation constituting a coded version of the input sound signal. In the calculation of the time-domain excitation contribution, the input sound signal may be processed in successive frames of the input sound signal and a number of sub-frames to be used in a current frame may be calculated.

Abstract: Provided are methods and systems for noise suppression within multiple time-frequency points of spectral representations. A multi-feature cluster tracker is used to track signal and noise sources and to predict signal versus noise dominance at each time-frequency point. Multiple features, such as binaural and monaural features, may be used for these purposes. A Gaussian mixture model (GMM) is developed and, in some embodiments, dynamically updated for distinguishing signal from noise and performing mask-based noise reduction. Each frequency band may use a different GMM or share a GMM with other frequency bands. A GMM may be combined from two models, with one trained to model time-frequency points in which the target dominates and another trained to model time-frequency points in which the noise dominates. Dynamic updates of a GMM may be performed using an expectation-maximization algorithm in an unsupervised fashion.

Abstract: The community-based generation of audio narrations for a text-based work leverages collaboration of a community of people to provide human-voiced audio readings. During the community-based generation, a collection of audio recordings for the text-based work may be collected from multiple human readers in a community. An audio recording for each section in the text-based work may be selected from the collection of audio recordings. The selected audio recordings may be then combined to produce an audio reading of at least a portion of the text-based work.

Abstract: An apparatus for determining a plurality of local center-of-gravity frequencies of a spectrum of an audio signal includes an offset determiner, a frequency determiner and an iteration controller. The offset determiner determines an offset frequency for each iteration start frequency of a plurality of iteration start frequencies based on the spectrum of the audio signal, wherein a number of discrete sample values of the spectrum is larger than a number of iteration start frequencies. The frequency determiner determines a new plurality of iteration start frequencies by increasing or reducing each iteration start frequency of the plurality of iteration start frequencies by the corresponding determined offset frequency. The iteration controller provides the new plurality of iteration start frequencies to the offset determiner for further iteration or provides the plurality of local center-of-gravity frequencies, if a predefined termination condition is fulfilled.

Abstract: An electronic device for coding a transient frame is described. The electronic device includes a processor and executable instructions stored in memory that is in electronic communication with the processor. The electronic device obtains a current transient frame. The electronic device also obtains a residual signal based on the current transient frame. Additionally, the electronic device determines a set of peak locations based on the residual signal. The electronic device further determines whether to use a first coding mode or a second coding mode for coding the current transient frame based on at least the set of peak locations. The electronic device also synthesizes an excitation based on the first coding mode if the first coding mode is determined. The electronic device also synthesizes an excitation based on the second coding mode if the second coding mode is determined.

Abstract: Apparatus, system and method for performing an action such as accessing supplementary data and/or executing software on a device capable of receiving multimedia are disclosed. After multimedia is received, a monitoring code is detected and a signature is extracted in response thereto from an audio portion of the multimedia. The ancillary code includes a plurality of code symbols arranged in a plurality of layers in a predetermined time period, and the signature is extracted from features of the audio of the multimedia. Supplementary data is accessed and/or software is executed using the detected code and/or signature.

Abstract: A signal processing method is provided. The signal processing method includes extracting a first signal having a first frequency band from a sum signal of a left signal and a right signal, generating a second signal having a second frequency band by using the first signal, generating a third signal by using the first signal and the second signal, and applying a gain, generated by using a rate of a center signal included in the sum signal, to the third signal.

Abstract: Method and system for tracking fundamental frequencies of pseudo-periodic signals in the presence of noise that include receiving a time-frequency representation of signals measured in a predefined environment; estimating and tracking a fundamental frequency of a respective pseudo-periodic signal at each time frame of the time-frequency representation by tracking detections of harmonious frequencies in the time-frequency representation over time; and outputting each respective estimated fundamental frequency associated with the pseudo-periodic signal of each respective time frame.

Abstract: From a plurality of received voice signals, a signal interval in which there is a talker collision between at least a first and a second voice signal is detected. A processor receives a positive detection result and processes, in response to this, at least one of the voice signals with the aim of making it perceptually distinguishable. A mixer mixes the voice signals to supply an output signal, wherein the processed signal(s) replaces the corresponding received signals. In example embodiments, signal content is shifted away from the talker collision in frequency or in time. The invention may be useful in a conferencing system.

Abstract: Method, system, and computer program product for voice transformation are provided. The method includes transforming a source speech using transformation parameters, and encoding information on the transformation parameters in an output speech using steganography, wherein the source speech can be reconstructed using the output speech and the information on the transformation parameters. A method for reconstructing voice transformation is also provided including: receiving an output speech of a voice transformation system wherein the output speech is transformed speech which has encoded information on the transformation parameters using steganography; extracting the information on the transformation parameters; and carrying out an inverse transformation of the output speech to obtain an approximation of an original source speech.

Abstract: A signal bandwidth extending apparatus including: a bandwidth extending section configured to extend a frequency bandwidth of a target signal, the target signal included in an input signal; a calculating section configured to calculate a degree of the target signal included in the input signal; and a controller configured to change a method of extending the frequency bandwidth by the bandwidth extending section according to a result of the calculating section.

Abstract: A method for decoding an audio signal in a decoder having a CELP-based decoder element including a fixed codebook component, at least one pitch period value, and a first decoder output, wherein a bandwidth of the audio signal extends beyond a bandwidth of the CELP-based decoder element. The method includes obtaining an up-sampled fixed codebook signal by up-sampling the fixed codebook component to a higher sample rate, obtaining an up-sampled excitation signal based on the up-sampled fixed codebook signal and an up-sampled pitch period value, and obtaining a composite output signal based on the up-sampled excitation signal and an output signal of the CELP-based decoder element, wherein the composite output signal includes a bandwidth portion that extends beyond a bandwidth of the CELP-based decoder element.

Abstract: A method and apparatus for transcoding audio data. The method includes determining if AAC joint stereo exists, running a reference AC-3 rematrixing when the AAC joint stereo does not exist, when AAC joint stereo does exist, enabling rematrixing when the number of corresponding AAC bands is greater than half the size of the band, otherwise, running reference AC-3 rematrixing.

Abstract: In a speech recognition device and a speech recognition method, a key phrase containing at least one key word is received. The speech recognition method comprises steps: receiving a sound source signal of a key word and generating a plurality of audio signals; transforming the audio signals into a plurality of frequency signals; receiving the frequency signals to obtain a space-frequency spectrum and an angular estimation value thereof; receiving the space-frequency spectrum to define and output at least one spatial eigenparameter, and using the angular estimation value and the frequency signals to perform spotting and evaluation and outputting a Bhattacharyya distance; and receiving the spatial eigenparameter and the Bhattacharyya distance and using corresponding thresholds to determine correctness of the key phrase. Thereby this invention robustly achieves high speech recognition rate under very low SNR conditions.

Abstract: A method for coding and decoding an audio signal or speech signal and an apparatus adopting the method are provided.

Abstract: Systems and methods are provided for acquiring a smooth spectrum of speech signals. For example, linear-spectrum-pairs (LSP) parameters of one or more speech signals to be processed are acquired; one or more first cosine values of the LSP parameters are calculated; one or more second cosine values are calculated for one or more predetermined frequency points; one or more first smooth spectrum values of the one or more predetermined frequency points are calculated based on at least information associated with the first cosine values of the LSP parameters and the second cosine values of the predetermined frequency points; and a smooth spectrum of the speech signals is generated based on at least information associated with the first smooth spectrum values of the predetermined frequency points.

Abstract: A mute setting is automatically set based on a speech detection result for acoustic signals received by a device. A device detects the speech based on a variety of cues from acoustic signals received using one or more microphones. If speech is detected within one or more frames, a mute setting may be automatically turned off. If speech is not detected, a mute setting may be automatically turned on. A mute setting may remain on as long as speech is not detected within the received acoustic signals. A varying delay may be implemented to help avoid false detections. The delay may be utilized during a mute-on state, and gradually removed during a transition from a mute-on state to a mute-off state.

Abstract: The present invention relates to blind source separation. More specifically certain embodiments relate to the blind source separation using frequency domain processes. Aspects of the invention relate to methods and systems for receiving a set of frequency-domain first signals, and then separating the set of frequency-domain first signals into a set of frequency-domain second signals. The frequency-domain second signals may have a set of separated frequency-domain second signal elements corresponding to individual frequencies wherein each frequency-domain second signal element is assigned an identifier. The identifier may indicate which of the set of frequency-domain second signals includes the frequency-domain second signal element. Some aspects also include reordering the identifiers corresponding to at least one frequency to improve coherence of the frequency-domain second signals and to produce a set of frequency-domain third signals.

Abstract: Methods of, apparatuses for, and computer readable media having instructions thereon that when executed cause carrying out methods of determining and modifying the perceived loudness of a frequency domain audio signal where the frequency resolution, and corresponding temporal coverage of the frequency domain information is not constant. The frequency (and thus temporal) resolution of the perceived loudness processing is maintained constant at the longest block size. One method includes a block combiner and a loudness modification interpolator.

Abstract: In accordance with an embodiment, a method of generating an encoded audio signal, the method includes estimating a time-frequency energy of an input audio signal from a time-frequency filter bank, computing a global variance of the time-frequency energy, determining a post-processing method according to the global variance, and transmitting an encoded representation of the input audio signal along with an indication of the determined post-processing method.

Abstract: System and methods for characterizing interest points within a fingerprint are disclosed herein. The systems include generating a set of interest points and an anchor point related to an audio sample. A quantized absolute frequency of an anchor point can be calculated and used to calculate a set of quantized ratios. A fingerprint can then be generated based upon the set of quantized ratios and used in comparison to reference fingerprints to identify the audio sample. The disclosed systems and methods provide for an audio matching system robust to pitch-shift distortion by using quantized ratios within fingerprints rather than solely using absolute frequencies of interest points. Thus, the disclosed system and methods result in more accurate audio identification.

Abstract: A method and apparatus to evaluate a quality of an audio signal, in which the number of effective channels is determined for each of a reference signal of a current frame and a test signal indicative of the reference signal that has passed through an audio codec, and an audio quality evaluation score of the current frame is calculated by evaluating an audio quality of the current frame based on the determined number of effective channels for each of the reference signal and the test signal by means of a predetermined evaluator.

Abstract: A method for decoding an audio signal having a bandwidth that extends beyond a bandwidth of a CELP excitation signal in an audio decoder including a CELP-based decoder element. The method includes obtaining a second excitation signal having an audio bandwidth extending beyond the audio bandwidth of the CELP excitation signal, obtaining a set of signals by filtering the second excitation signal with a set of bandpass filters, scaling the set of signals using a set of energy-based parameters, and obtaining a composite output signal by combining the scaled set of signals with a signal based on the audio signal decoded by the CELP-based decoder element.

Abstract: A signal processing apparatus feeding a frame of a signal in frequency domain of a reception voice signal into a sound echo canceler includes a first reception section for receiving frames of the reception voice signal in frequency domain before having a rate-of-speech change process applied; a second reception section for receiving frames of a signal in time domain having the rate-of-speech change process applied by units of frames; and a frequency-domain frame synthesis section for synthesizing a frame of the signal in frequency domain of the reception voice signal based on the signal in time domain having the rate-of-speech change process applied at a frame currently being processed by the signal processing apparatus, and a frame of the reception voice signal in frequency domain corresponding to the signal in time domain having the rate-of-speech change process applied.

Abstract: A system enhances speech by detecting a speaker's utterance through a first microphone positioned a first distance from a source of interference. A second microphone may detect the speaker's utterance at a different position. A monitoring device may estimate the power level of a first microphone signal. A synthesizer may synthesize part of the first microphone signal by processing the second microphone signal. The synthesis may occur when power level is below a predetermined level.

Abstract: Audio decoder and method therein for supporting bandwidth extension (BWE) of a received signal. The method involves receiving a first signal representing the lower frequency spectrum of a segment of an original audio signal; receiving a second signal, being a BWE signal, representing a higher frequency spectrum of the segment of the original audio signal. The method further comprises determining a degree of voicing in the lower frequency spectrum of the audio signal, based on the received first signal; and selecting a spectral tilt adaptation filter, out of at least two spectral tilt adaptation filters having different spectral attenuation characteristics, based on the determined degree of voicing. The selected spectral tilt adaptation filter is then applied on the received second signal. Thus, a differentiation of spectral tilt in the higher frequency spectrum of a reconstructed audio signal, based on lower frequency spectrum characteristics of the original audio signal is enabled.

Abstract: A framing method and apparatus are disclosed to overcome inconsistency of gains between sub-frames caused by simple average framing in the prior art. The method includes: obtaining the Linear Prediction Coding (LPC) order and the pitch of the signal; removing the samples inapplicable to Long-Term Prediction (LTP) synthesis according to the LPC prediction order and the pitch; and splitting the remaining samples of the signal into several sub-frames. The technical solution under the present invention is applicable to the multimedia speech coding field.

Abstract: A system and method for representing quasi-periodic waveforms, for example, representing a plurality of limited decompositions of the quasi-periodic waveform. Each decomposition includes a first and second amplitude value and at least one time value. In some embodiments, each of the decompositions is phase adjusted such that the arithmetic sum of the plurality of limited decompositions reconstructs the quasi-periodic waveform. Data-structure attributes are created and used to reconstruct the quasi-periodic waveform. Features of the quasi-periodic wave are tracked using pattern-recognition techniques. The fundamental rate of the signal (e.g., heartbeat) can vary widely, for example by a factor of 2-3 or more from the lowest to highest frequency. To get quarter-phase representations of a component (e.g.

Abstract: An acoustic signal processing system and method. In accordance with an embodiment, the acoustic signal processing system includes an adaptive filter that filters a signal from a frequency band reservation module and generates a filter signal that is received by a subtractor. The subtractor generates an error signal that is used by a double-talk indicator module to generate a control signal that indicates the presence of double-talk.

Abstract: The present invention is a means to provide a user interface that will naturally cause a person to speak at a normal talking volume. It is based on a mechanism whereby the user's speech is compared to a threshold to determine if the user is speaking too loudly and provides feedback to the user. This mechanism could be incorporated into a headset, a cell phone, a smartphone, or into other communication devices. It is useful for operation with or without a headset.

Abstract: An audio encoding apparatus capable of reducing the bit rate even if a codebook having a larger codebook number is selected in a split multi-rate lattice vector quantization is provided. Sub-vector determining unit (121) determines, in the spectrum of an input signal having been divided into a predetermined number of sub-vectors, a sub-vector using the largest number of bits. Positional information encoding unit (122) encodes the positional information of the determined sub-vector. Codebook indication value estimating unit (124) estimates a number of used bits for a codebook indication value of the largest number of used bits by use of the (N?1) other codebook indication values, and generates a number-of-used-bits estimation value. Difference calculating unit (125) calculates a difference by subtracting the number-of-used-bits estimation value from the actual value of the codebook indication value of the largest number of used bits. Difference encoding unit (126) encodes the difference information.

Abstract: A special rendering mode for the first few seconds of play out of multimedia data minimizes the delay caused by pre-buffering of data packets in multimedia streaming applications. Instead of pre-buffering all incoming data packets until a certain threshold is reached, the streaming application starts playing out some of the data packets immediately after the arrival of the first data packet. Immediate play out of the first data packet, for example, results in minimum delay between channel selection and perception, thereby allowing a user to quickly scan through all available channels to quickly get a notion of the content. The immediate play out is done at a reduced speed.

Abstract: This encoding device (100) is provided with: a CELP encoding unit (102) that decodes CELP encoded data resulting from CELP encoding an input signal, generating a CELP decoded signal; a transform encoding unit (106) that generates a decoded signal spectrum by decoding transform encoded data resulting from using the spectrum of the input signal and the suppression spectrum of suppressing using a first suppression factor to transform encode the amplitude of the spectrum of the CELP decoded signal, and that outputs an index of the transform encoded frequency component; a pulse index recording unit (107) that forms and records an array using the index; and a CELP component suppression unit (109) that uses a second suppression factor and the array to suppress the amplitude of the spectrum resulting from adding the decoded signal spectrum and the suppression spectrum.

Abstract: A sound source generating unit 101 generates from a narrowband audio signal not passing through noise suppression a sound source signal including a fine structure of a band to be restored. On the other hand, a noise suppressing unit 102 performs noise suppression of the narrowband audio signal and a spectral envelope estimating unit 103 estimates an spectral envelope of the band to be restored. A signal synthesizing unit 104 generates a pseudo-audio signal by combining the sound source signal and the spectral envelope, and the band-pass filter unit 105 passes the pseudo-audio signal of the band to be restored, and the signal addition unit 106 generates a broadband audio restoration signal by adding the pseudo-audio signal of the band to be restored to the narrowband audio signal.