Abstract: Disclosed are a vector quantization device and others capable of adaptively adjusting a vector space of a code vector for quantization of a second stage by using a quantization result of a first stage and improving the quantization accuracy.

Abstract: A DTMF signal processing apparatus of the present invention comprises a data divider unit, a DTMF signal component analyzer unit, a weighting processing unit, a buffer, a DTMF signal erasure determination unit, and a DTMF signal erasure processing unit. The data divider unit divides speech data into a plurality of divided speech data, and the DTMF signal component analyzer unit analyzes whether or not the divided speech data has a DTMF signal component. The weighting processing unit applies a weighting value to divided speech data analyzed at this time and stores the resultant speech data in the buffer, and also applies a weighting value to past divided speech data previously stored in the buffer when the result of the analysis indicates that the analyzed divided speech data has the DTMF signal component. The DTMF signal erasure determination unit determines based on the weighting value whether or not to erase the divided speech data stored in the buffer.

Abstract: A characteristic thumbprint is extracted from a data signal, the thumbprint based on statistics relating to the data signal. The data signal can be compared indirectly by matching this thumbprint against one or more reference thumbprints. The data signal may be any type of signal, including streaming digitized audio or obtained from static files. A database may contain a number of these characteristic thumbprints, and the database can be searched for a particular thumbprint.

Abstract: Provided is a method for converting a dimension of a vector. The vector dimension conversion method for vector quantization includes the steps of: extracting a specific parameter having a pitch period from an input speech signal and then generating a vector of a dimension that varies according to the pitch period; dividing an entire frequency domain of the generated vector of the variable dimension into at least two frequency domains; and converting the vector of the variable dimension into vectors of mutually different fixed dimensions according to the divided frequency domains. Thereby, not only an error due to the vector dimension conversion is suppressed but codebook memory required for the vector quantization is effectively reduced.

Abstract: An apparatus for providing voice conversion using temporal dynamic features includes a feature extractor and a transformation element. The feature extractor may be configured to extract dynamic feature vectors from source speech. The transformation element may be in communication with the feature extractor and configured to apply a first conversion function to a signal including the extracted dynamic feature vectors to produce converted dynamic feature vectors. The first conversion function may have been trained using at least dynamic feature data associated with training source speech and training target speech. The transformation element may be further configured to produce converted speech based on an output of applying the first conversion function.

Abstract: An audio encoder performs entropy encoding of audio data. For example, an audio encoder determines whether a first code table in a group of plural code tables contains a code representing a first vector. If it does, the code is used, and otherwise the escape code from the first code table is used and the first vector is encoded using, at least in part, a second code table. An audio decoder performs corresponding entropy decoding.

Abstract: An audio encoder performs entropy encoding of audio data. For example, an audio encoder determines a Huffman code from a Huffman code table to use for encoding a vector of audio data symbols, where the determining is based on a sum of values of the audio data symbols. An audio decoder performs corresponding entropy decoding.

Abstract: There is provided a vector conversion device for converting a reference vector used for quantization of an input vector so as to improve a signal quality including audio. In this vector conversion device, a vector quantization unit (902) acquires a number corresponding to a decoded LPC parameter of a narrow band from all the code vectors stored in a code book (903). A vector dequantization unit (904) references the number of the code vector obtained by the vector quantization unit (902) and selects a code vector from the code book (905). A conversion unit (906) performs calculation by using a sampling-adjusted decoded LPC parameter obtained from an up-sampling unit (901) and a code vector obtained from the vector dequantization unit (904), thereby obtaining a decoded LPC parameter of a broad band.

Abstract: A novel system for speech recognition uses differential cepstra over time frames as acoustic features, together with the traditional static cepstral features, for hidden trajectory modeling, and provides greater accuracy and performance in automatic speech recognition. According to one illustrative embodiment, an automatic speech recognition method includes receiving a speech input, generating an interpretation of the speech, and providing an output based at least in part on the interpretation of the speech input. The interpretation of the speech uses hidden trajectory modeling with observation vectors that are based on cepstra and on differential cepstra derived from the cepstra. A method is developed that can automatically train the hidden trajectory model's parameters that are corresponding to the components of the differential cepstra in the full acoustic feature vectors.

Abstract: Disclosed are systems, methods, and computer readable media for performing speech recognition. The method embodiment comprises selecting a codebook from a plurality of codebooks with a minimal acoustic distance to a received speech sample, the plurality of codebooks generated by a process of (a) computing a vocal tract length for a each of a plurality of speakers, (b) for each of the plurality of speakers, clustering speech vectors, and (c) creating a codebook for each speaker, the codebook containing entries for the respective speaker's vocal tract length, speech vectors, and an optional vector weight for each speech vector, (2) applying the respective vocal tract length associated with the selected codebook to normalize the received speech sample for use in speech recognition, and (3) recognizing the received speech sample based on the respective vocal tract length associated with the selected codebook.

Abstract: A vector quantization apparatus performs coding of a linear predictive coding coefficient by multi-stage vector quantization. A first codebook and a second codebook store code vectors, and a storing section stores scalars. A first quantizing section extracts a first code vector stored in the first codebook and performs first stage quantization for quantizing a target vector using the first code vector. A second quantizing section extracts a second code vector stored in the second codebook, calculates a third code vector by multiplying the second code vector and one of the scalars stored in the storing section, performs distance calculation using the target vector, the first code vector and the third code vector, and performs second stage quantization for quantizing the target vector using a result of the distance calculation. Each scalar stored in the storing section is associated with at least one of the vectors stored in the first codebook.

Abstract: A method and apparatus for prediction in a speech-coding system is provided herein. The method of a 1st order long-term predictor (LTP) filter, using a sub-sample resolution delay, is extended to a multi-tap LTP filter, or, viewed from another vantage point, the conventional integer-sample resolution multi-tap LTP filter is extended to use sub-sample resolution delay. This novel formulation of a multi-tap LTP filter offers a number of advantages over the prior-art LTP filter configurations. Particularly, defining the lag with sub-sample resolution makes it possible to explicitly model the delay values that have a fractional component, within the limits of resolution of the over-sampling factor used by the interpolation filter. The coefficients of such a multi-tap LTP filter are thus largely freed from modeling the effect of delays that have a fractional component.

Abstract: An encoding device (200) includes an MDCT unit (202) that transforms an input signal in a time domain into a frequency spectrum including a lower frequency spectrum, a BWE encoding unit (204) that generates extension data which specifies a higher frequency spectrum at a higher frequency than the lower frequency spectrum, and an encoded data stream generating unit (205) that encodes to output the lower frequency spectrum obtained by the MDCT unit (202) and the extension data obtained by the BWE encoding unit (204). The BWE encoding unit (204) generates as the extension data (i) a first parameter which specifies a lower subband which is to be copied as the higher frequency spectrum from among a plurality of the lower subbands which form the lower frequency spectrum obtained by the MDCT unit (202) and (ii) a second parameter which specifies a gain of the lower subband after being copied.

Abstract: The present invention relates to a gain quantization method and device for implementation in a technique for coding a sampled sound signal processed, during coding, by successive frames of L samples, wherein each frame is divided into a number of subframes and each subframe comprises a number N of samples, where N<L. In the gain quantization method and device, an initial pitch gain is calculated based on a number f of subframes, a portion of a gain quantization codebook is selected in relation to the initial pitch gain, and pitch and fixed-codebook gains are jointly quantized. This joint quantization of the pitch and fixed-codebook gains comprises, for the number f of subframes, searching the gain quantization codebook in relation to a search criterion. The codebook search is restricted to the selected portion of the gain quantization codebook and an index of the selected portion of the gain quantization codebook best meeting the search criterion is found.

Abstract: The present invention relates to a method of coding a signal, in particular an audio or speech signal, wherein a codebook comprising k code vectors is provided for vector quantization of a signal vector representing a set of signal values of said signal(s), and wherein an optimal code vector of said codebook is determined by performing a codebook search. Parallelism is employed to accelerate the coding procedure. In particular, the codebook search is highly parallelised.

Abstract: A system, method and computer program product for classification of an analog electrical signal using statistical models of training data. A technique is described to quantize the analog electrical signal in a manner which maximizes the compression of the signal while simultaneously minimizing the diminution in the ability to classify the compressed signal. These goals are achieved by utilizing a quantizer designed to minimize the loss in a power of the log-likelihood ratio. A further technique is described to enhance the quantization process by optimally allocating a number of bits for each dimension of the quantized feature vector subject to a maximum number of bits available across all dimensions.

Abstract: An encoder, decoder, encoding method, and decoding method enabling acquisition of high-quality decoded signal in scalable encoding of an original signal in first and second layers even if the second or upper layer section performs low bit-rate encoding. In the encoder, a spectrum residue shape codebook (305) stores candidates of spectrum residue shape vectors, a spectrum residue gain codebook (307) stores candidates of spectrum residue gains, and a spectrum residue shape vector and a spectrum residue gain are sequentially outputted from the candidates according to the instruction from a search section (306). A multiplier (308) multiplies a candidate of the spectrum residue shape vector by a candidate of the spectrum residue gain and outputs the result to a filtering section (303).

Abstract: It is an object to disclose an adaptive sound source vector quantizing device, etc. that can be configured to improve quantizing accuracy in adaptive sound source vector quantization to be carried out for every sub-frame.

Abstract: An audio encoder/decoder performs band partitioning for vector quantization encoding of spectral holes and missing high frequencies that result from quantization when encoding at low bit rates. The encoder/decoder determines a band structure for spectral holes based on two threshold parameters: a minimum hole size threshold and a maximum band size threshold. Spectral holes wider than the minimum hole size threshold are partitioned evenly into bands not exceeding the maximum band size threshold in size. Such hole filling bands are configured up to a preset number of hole filling bands. The bands for missing high frequencies are then configured by dividing the high frequency region into bands having binary-increasing, linearly-increasing or arbitrarily-configured band sizes up to a maximum overall number of bands.

Abstract: A vector quantization codebook search method and apparatus use support vector machines (“SVMs”) to compute a hyperplane, where the hyperplane is used to separate codebook elements into a plurality of bins. During execution, a controller determines which of the plurality of bins contains a desired codebook element, and then searches the determined bin. Codebook search complexity is reduced and an exhaustive codebook search is selectively avoided.

Abstract: A method for coding speech or other generic signals includes dividing a speech signal into a plurality of frames, and dividing at least one of the plurality of frames into at least two subframe units. A search for a fixed codebook contribution and an adaptive codebook contribution for subframe units is conducted. At least one subframe unit is selected to be coded without the fixed codebook contribution. The encoder may iteratively arrange and encode subframes differently for the same frame, and select for transmission that arrangement that minimizes an error measure across the frame. Various embodiments are shown, as are embodied computer programs, a decoder, and a communication system.

Abstract: A high quality speech is reproduced with a small data amount in speech coding and decoding for performing compression coding and decoding of a speech signal to a digital signal. In speech coding method according to a code-excited linear prediction (CELP) speech coding, a noise level of a speech in a concerning coding period is evaluated by using a code or coding result of at least one of spectrum information, power information, and pitch information, and various excitation codebooks are used based on an evaluation result.

Abstract: Coding of an audio signal (x) represented by a respective set of sampled signal values (x(t)) for each of a plurality of sequential time segments is disclosed. The sampled signal values are analyzed to determine one or more sinusoidal components for each of the plurality of sequential segments. The sinusoidal components are linked across a plurality of sequential segments to provide sinusoidal tracks, where each track comprises a number of frames. An encoded signal (AS) is generated, including sinusoidal codes (Cs) comprising a representation level (r) for each frame or including sinusoidal codes (Cs) where some of these codes comprise a phase (?), a frequency (?) and a quantization table (Q) for a given frame when the given frame is designated as a random-access frame. The invention allows random access in a track while avoiding long adaptation of the quantization accuracy in a quantizer and/or the need for a large bit stream while still maintaining improved audio quality.

Abstract: An apparatus and method for encoding/decoding a speech signal which determines a variable bit rate based on reserved bits obtained from a target bit rate, is provided. The variable bit rate is determined based on a source feature of the speech signal and the reserved bits is obtained based on the target bit rate. The apparatus for encoding the speech signal may include a linear predictive (LP) analysis unit/quantization unit to determine an immittance spectral frequencies (ISF) index, a closed loop pitch search unit, a fixed codebook search unit, a gain vector quantization (VQ) unit to determine a gain vector quantization (VQ) index, and a bit rate control unit to control at least two indexes of the ISF index, the pitch index, the code index, and the gain VQ index to be encoded to be variable bit rates based on a source feature of a speech signal and the reserved bits.

Abstract: Methods and apparatus are presented for reducing the number of bits needed to represent an excitation waveform. An acoustic signal in an analysis frame is analyzed to determine whether it is a band-limited signal. A sub-sampled sparse codebook is used to generate the excitation waveform if the acoustic signal is a band-limited signal. The sub-sampled sparse codebook is generated by decimating permissible pulse locations from the codebook track in accordance with the frequency characteristic of the acoustic signal.

Abstract: A voice and musical tone coding apparatus is provided that can perform high-quality coding by executing vector quantization taking the characteristics of human hearing into consideration. In this voice and musical tone coding apparatus, a quadrature transformation processing section (201) converts a voice and musical tone signal from time components to frequency components. An auditory masking characteristic value calculation section (203) finds an auditory masking characteristic value from a voice and musical tone signal. A vector quantization section (202) performs vector quantization changing a calculation method of a distance between a code vector found from a preset codebook and a frequency component based on an auditory masking characteristic value.

Abstract: The invention relates to compression coding and/or decoding of digital signals, in particular by vector variable-rate quantisation defining a variable resolution. For this purpose an impulsion dictionary comprises: for a given dimension, increasing resolution dictionaries imbricated into each other and, for a given dimension, a union of: a totality (D?i<N>) of code-vectors produced, by inserting elements taken in a final set (A) into smaller dimension code-vectors according to a final set of predetermined insertion rules (F1) and a second totality of code-vectors (Y?) which are not obtainable by insertion into the smaller dimension code vectors according to said set of the insertion rules.

Abstract: Provided is an adaptive sound source vector quantization device which can always perform a pitch cycle search with a resolution appropriate for any section of the pitch cycle search range of a second sub-frame when a pitch cycle search range of the second sub-frame changes in accordance with a pitch cycle of a first sub-frame. The device includes a first pitch cycle instruction unit (111), a search range calculation unit (112), and a second pitch cycle instruction unit (113). The first pitch cycle instruction unit (111) successively instructs pitch cycle search candidates in a predetermined search range having a search resolution which transits over a predetermined pitch cycle candidate for the first sub-frame. The search range calculation unit (112) calculates a predetermined range before and after the pitch cycle of the first sub-frame as the pitch cycle search range for the second sub-frame, if the predetermined range includes the predetermined pitch cycle search candidate.

Abstract: A method for coding data, includes: grouping data into frames; classifying the frames into classes; for each class, transforming the frames belonging to the class into filter parameter vectors, which are extracted from the frames by applying a first mathematical transformation; for each class, computing a filter codebook based on the filter parameter vectors belonging to the class; segmenting each frame into subframes; for each class, transforming the subframes belonging to the class into source parameter vectors, which are extracted from the subframes by applying a second mathematical transformation based on the filter codebook computed for the corresponding class; for each class, computing a source codebook based on the source parameter vectors belonging to the class; and coding the data based on the computed filter and source codebooks.

Abstract: Provided is an encoding device which can obtain a sound quality preferable for auditory sense even if the number of information bits is small. The encoding device includes a shape quantization unit (111) having: a section search unit (121) which searches for a pulse for each of bands into which a predetermined search section is divided; and a whole search unit (122) which performs search for a pulse over the entire search section. The shape of an input spectrum is quantized by a small number of pulse positions and polarities. A gain quantization unit (112) calculates a gain of the pulse searched by the shape quantization unit (111) and quantizes the gain for each of the bands.

Abstract: Disclosed are an audio encoding device and an audio decoding device which reduce degradation of subjective quality of a decoding signal caused by power mismatch of a decoding signal which is generated by a concealing process upon disappearance of a frame. When a frame is lost, a past encoding parameter is used to obtain a concealed LPC of the current frame and a concealed sound source parameter. A normal CELP decoding is performed from the obtained concealed sound source parameter. Correction is performed by using a conceal parameter on the obtained concealed LPC and the concealed sound source signal. The power of the corrected concealed sound source signal is adjusted to match a reference sound source power. A filter gain of the synthesis filter is adjusted so as to adjust the power of a decoded sound signal to the power of a decoded sound signal during an error-free state.

Abstract: Provided is an audio encoding device which performs a closed loop search of a gain and a sound source vector without significantly increasing the calculation amount as compared to an open loop search. In the audio encoding device, firstly, a first parameter decision unit (121) performs a sound source search by an adaptive sound source codebook and then a second parameter decision unit (122) simultaneously performs by a closed loop, the sound source and the gain search by using a fixed sound source codebook. More specifically, for a combination of a fixed sound source vector and gain, the sum of a value obtained by multiplying a candidate fixed sound source vector by a candidate gain and a value obtained by multiplying an adaptive sound source vector by a candidate gain is subjected to a combination filter formed by a filter coefficient based on a quantization linear prediction coefficient so as to generate a combined signal.

Abstract: A method of performing quantization in an audio encoder includes determining a number of bits available in a frame of encoded audio data. Determinations are also made for the maximum transform coefficient value and a distribution of transform coefficient values across the transform coefficient spectrum being encoded. A an estimate for an initial quantization step value is determined from the number of available bits in the frame, the maximum transform coefficient value, and the distribution of coefficient values across the coefficient spectrum.

Abstract: Coding of an audio signal represented by a respective set of sampled signal values for each of a plurality of sequential segments is disclosed. The sampled signal values are analyzed (40) to determine one or more sinusoidal components for each of the plurality of sequential segments. The sinusoidal components are linked (42) across a plurality of sequential segments to provide sinusoidal tracks. For each sinusoidal track, a phase comprising a generally monotonically changing value is determined and an encoded audio stream including sinusoidal codes (r) representing said phase is generated (46).

Abstract: A speech compression apparatus including: a first band-transform unit transforming a wideband speech signal to a narrowband low-band speech signal; a narrowband speech compressor compressing the narrowband low-band speech signal and outputting a result of the compressing as a low-band speech packet; a decompression unit decompressing the low-band speech packet and obtaining a decompressed wideband low-band speech signal; an error detection unit detecting an error signal that corresponds to a difference between the wideband speech signal and the decompressed wideband low-band speech signal; and a high-band speech compression unit compressing the error signal and a high-band speech signal of the wideband speech signal and outputting the result of the compressing as a high-band speech packet.

Abstract: A method comprises identifying a component k of a codevector from a codebook C having one or more codevectors, the component k introducing highest variance for an input vector; allowing ordering of codevectors in the codebook C; and searching for a best match vector for the input vector using ordered codevectors.

Abstract: Provided is a voice encoding device which can accurately encode a spectrum shape of a signal having a strong tonality such as a vowel. The device includes: a sub-band constituting unit (151) which divides a first layer error conversion coefficient to be encoded into M sub-bands so as to generate M sub-band conversion coefficients; a shape vector encoding unit (152) which performs encoding on each of the M sub-band conversion coefficient so as to obtain M shape encoded information and calculates a target gain of each of the M sub-band conversion coefficients; a gain vector forming unit (153) which forms one gain vector by using M target gains; a gain vector encoding unit (154) which encodes the gain vector so as to obtain gain encoded information; and a multiplexing section unit (155) which multiplexes the shape encoded information with the gain encoded information.

Abstract: A method and an apparatus for encoding digital audio data with reduced bit rates, the apparatus having a provider of psycho-acoustically quantized digital audio data with a bit rate being higher than the reduced bit rate. The apparatus further has an identifier for identifying a frequency band according to a selection criterion, the selection criterion being such that an impact on the quality of the digital audio data when the data in the identified frequency band is replaced by generated noise is smaller than the impact on the quality of the digital audio data, which would arise when the data in a different frequency band is replaced by generated noise. The apparatus further has a replacer for replacing data in the identified frequency band of the digital audio data by a noise synthesis parameter.

Abstract: An aspect of the present invention takes advantage of the fact that the coordinates of fixed points do not change, and thus the energy (sum of squares of the coordinates defining the vector) of each fixed point is computed and stored. The energy of each variable input point may also be computed. The distance between each pair of fixed and input points is computed based on the respective energies and the dot product.

Abstract: A low bit rate digital audio coding system includes an encoder which assigns codebooks to groups of quantization indexes based on their local properties resulting in codebook application ranges that are independent of block quantization boundaries. The invention also incorporates a resolution filter bank, or a tri-mode resolution filter bank, which is selectively switchable between high and low frequency resolution modes or high, low and intermediate modes such as when detecting transient in a frame. The result is a multichannel audio signal having a significantly lower bit rate for efficient transmission or storage. The decoder is essentially an inverse of the structure and methods of the encoder, and results in a reproduced audio signal that cannot be audibly distinguished from the original signal.

Abstract: A block-constrained Trellis coded quantization (TCQ) method and a method and apparatus for quantizing line spectral frequency (LSF) parameters employing the same in a speech coding system wherein the LSF coefficient quantizing method includes: removing the direct current (DC) component in an input LSF coefficient vector; generating a first prediction error vector by performing inter-frame and intra-frame prediction for the LSF coefficient vector, in which the DC component is removed, quantizing the first prediction error vector by using the BC-TCQ algorithm, and by performing intra-frame and inter-frame prediction compensation, generating a quantized first LSF coefficient vector; generating a second prediction error vector by performing intra-frame prediction for the LSF coefficient vector, in which the DC component is removed, quantizing the second prediction error vector by using the BC-TCQ algorithm, and then, by performing intra-frame prediction compensation, generating a quantized second LSF coefficient

Abstract: A method of deriving a codebook including deriving a first codeword Pl in Euclidean coordinates, transforming the first codeword into eigen-coordinates, applying a Hochwald construction to the first codeword in eigen-coordinates to derive a plurality of codewords in eigen-coordinates and transforming the plurality of codewords in eigen-coordinates into a plurality of codewords in Euclidean coordinates to form the codebook.

Abstract: A vector quantizing device for dividing a sequence of vectors and quantizing them with an enhanced performance of vector quantization by using information on the correlation between the high and low order that the vector sequence has. The vector quantizing device (100) creates a predicted vector by prediction using a first quantization divided vector, creates the difference between the divided vector and the predicted vector as a predicted residual vector, and determines a second code by converting the predicted residual vector into a quantized vector. A vector dequantizing device (150) creates a predated vector by prediction using a first quantization divided vector, creates a second quantization divided vector by adding the predicted vector and the predicted residual vector, and creates a quantized vector by connecting the first and second quantization divided vectors.

Abstract: A speech compression apparatus including: a first band-transform unit transforming a wideband speech signal to a narrowband low-band speech signal; a narrowband speech compressor compressing the narrowband low-band speech signal and outputting a result of the compressing as a low-band speech packet; a decompression unit decompressing the low-band speech packet and obtaining a decompressed wideband low-band speech signal; an error detection unit detecting an error signal that corresponds to a difference between the wideband speech signal and the decompressed wideband low-band speech signal; and a high-band speech compression unit compressing the error signal and a high-band speech signal of the wideband speech signal and outputting the result of the compressing as a high-band speech packet.

Abstract: A quantizer for quantization of a vector comprises a codevector generator that generates a set of candidate codevectors and a memory for storing an illegal space definition representing illegal vectors. The quantizer also includes a legal status tester that determines legal candidate codevectors among the set of candidate codevectors using the illegal space definition, and a codevector selector that determines a best legal candidate codevector among the one or more legal candidate codevectors. The vector includes parameters relating to a speech and/or audio signal, such as Line Spectral Frequencies (LSFs).

Abstract: An audio signal encoding apparatus includes a frame dividing unit (1), an auditory psychological arithmetic unit (2), a filter bank unit (3), a scale factor calculation unit (4) which weights the spectra in the respective frequency bands by an arithmetic result of the auditory psychological arithmetic unit (2), a quantization step determination unit (7) which determines a quantization step of the entire frame prior to spectrum quantization by subtracting an information size of all quantized spectra from an auditory information size of all the weighted spectra before quantization, and multiplying the difference by a coefficient obtained from a step width of a quantization coarseness, a spectrum quantization unit (8), and a bit shaping unit (9) which outputs a bitstream obtained by shaping quantized spectra. The quantization step determination unit predicts the information size of all the quantized spectra based on a bit size assigned to a frame to be encoded.

Abstract: A method and apparatus for searching fixed codebook are provided. The method includes: obtaining a basic codebook which comprises position information of N pulses on M tracks, wherein N and M are positive integers; choosing n pulses as search pulses, wherein the n pulses are parts of the N pulses and n is a positive integer smaller than N; and replacing position information of the n search pulses respectively with other position information on the tracks to obtain a searched codebook; executing the search process for K times, wherein K is a positive integer larger than or equal to 2, at least two or more search pulses are chosen in one of the K search processes , and the chosen search pulses vary in each of the K search processes; and obtaining an optimal codebook from the basic codebook and the searched codebook according to a preset criterion.

Abstract: A code excited linear prediction speech decoder includes an adaptive codebook configured to generate an adaptive code vector. The decoder also includes a random codebook configured to generate a random code vector. The decoder also includes a synthesis filter that receives a signal based on said adaptive code vector and said random code vector, and that is configured to perform linear prediction coefficient synthesis on said signal. The random codebook includes a pulse vector providing system configured to provide a pulse vector having a signed unit pulse. The random codebook also includes a comparing system configured to compare a value of adaptive codebook gain with a preset threshold value. The random codebook further includes a selecting system configured to select a dispersion pattern from a plurality of dispersion patterns stored in a memory in accordance with a result of said comparison.

Abstract: This invention relates to a method, a device and a software application product for N-level quantization of vectors, wherein N is selectable prior to said quantization from a set of at least two pre-defined values that are smaller than or equal to a pre-defined maximum number of levels M. A reproduction vector for each vector is selected from an N-level codebook of N reproduction vectors that are, for each N in said set of at least two pre-defined values, represented by the first N reproduction vectors of the same joint codebook of M reproduction vectors. The invention further relates to a method, a device and a software application product for retrieving reproduction vectors for vectors that have been N-level quantized, to a system for transferring representations of vectors, to a method, a device and a software application product for determining a joint codebook, and to such a joint codebook itself.

Abstract: An enhanced analysis-by-synthesis waveform interpolative speech coder able to operate at 2.8 kbps. Novel features include dual-predictive analysis-by-synthesis quantization of the slowly-evolving waveform, efficient parametrization of the rapidly-evolving waveform magnitude, and analysis-by-synthesis vector quantization of the rapidly evolving waveform parameter. Subjective quality tests indicate that it exceeds G.723.1 at 5.3 kbps, and of G.723.1 at 6.3 kbps.