WebCPC CPC COOPERATIVE PATENT CLASSIFICATION

G10L　SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING

G10L 13/00 Speech synthesis; Text to speech systems

G10L 13/02 ・Methods for producing synthetic speech; Speech synthesisers

G10L 13/027 ・・Concept to speech synthesisers; Generation of natural phrases from machine-based concepts (generation of parameters for speech synthesis out of text G10L 13/08)

G10L 13/033 ・・Voice editing, e.g. manipulating the voice of the synthesiser

G10L 13/0335 ・・・{Pitch control}

G10L 13/04 ・・Details of speech synthesis systems, e.g. synthesiser structure or memory management

G10L 13/043 ・・・{Synthesisers specially adapted to particular applications}

G10L 13/047 ・・・Architecture of speech synthesisers

G10L 13/06 ・Elementary speech units used in speech synthesisers; Concatenation rules

G10L 13/07 ・・Concatenation rules

G10L 13/08 ・Text analysis or generation of parameters for speech synthesis out of text, e.g. grapheme to phoneme translation, prosody generation or stress or intonation determination

G10L 13/086 ・・{Detection of language}

G10L 13/10 ・・Prosody rules derived from text; Stress or intonation

G10L 15/00 Speech recognition (G10L 17/00 takes precedence)

G10L 15/005 ・{Language recognition}

G10L 15/01 ・Assessment or evaluation of speech recognition systems

G10L 15/02 ・Feature extraction for speech recognition; Selection of recognition unit

G10L 15/04 ・Segmentation; Word boundary detection

G10L 15/05 ・・Word boundary detection

G10L 15/06 ・Creation of reference templates ; Training of speech recognition systems, e.g. adaptation to the characteristics of the speaker's voice (G10L 15/14 takes precedence)

G10L 15/063 ・・{Training}

G10L 15/065 ・・Adaptation

G10L 15/07 ・・・to the speaker

G10L 15/075 ・・・・{supervised, i.e. under machine guidance}

G10L 15/08 ・Speech classification or search

G10L 15/083 ・・{Recognition networks (G10L 15/142, G10L 15/16 take precedence)}

G10L 15/10 ・・using distance or distortion measures between unknown speech and reference templates

G10L 15/12 ・・using dynamic programming techniques, e.g. dynamic time warping [DTW]

G10L 15/14 ・・using statistical models, e.g. hidden Markov models [HMMs] (G10L 15/18 takes precedence)

G10L 15/142 ・・・{Hidden Markov Models [HMMs} ]

G10L 15/144 ・・・・{Training of HMMs}

G10L 15/146 ・・・・・{with insufficient amount of training data, e.g. state sharing, tying, deleted interpolation}

G10L 15/148 ・・・・[N: Duration modelling in HMMs, e.g. semi HMM, segmental models or transition probabilities

G10L 15/16 ・・using artificial neural networks

G10L 15/18 ・・using natural language modelling

G10L 15/1807 ・・・{using prosody or stress}

G10L 15/1815 ・・・{Semantic context, e.g. disambiguation of the recognition hypotheses based on word meaning}

G10L 15/1822 ・・・{Parsing for meaning understanding}

G10L 15/183 ・・・using context dependencies, e.g. language models

G10L 15/187 ・・・・Phonemic context, e.g. pronunciation rules, phonotactical constraints or phoneme n-grams

G10L 15/19 ・・・・Grammatical context, e.g. disambiguation of the recognition hypotheses based on word sequence rules

G10L 15/193 ・・・・・Formal grammars, e.g. finite state automata, context free grammars or word networks

G10L 15/197 ・・・・・Probabilistic grammars, e.g. word n-grams

G10L 15/20 ・Speech recognition techniques specially adapted for robustness in adverse environments, e.g. in noise, of stress induced speech (G10L 21/02 takes precedence)

G10L 15/22 ・Procedures used during a speech recognition process, e.g. man-machine dialogue

G10L 15/222 ・・{Barge in, i.e. overridable guidance for interrupting prompts}

G10L 15/24 ・Speech recognition using non-acoustical features

G10L 15/25 ・・using position of the lips, movement of the lips or face analysis

G10L 15/26 ・Speech to text systems (G10L 15/08 takes precedence)

G10L 15/265 ・・{Speech recognisers specially adapted for particular applications (devices for signalling identity of wanted subscriber in a telephonic communication equipment controlled by voice recognition H04M 1/271; speech interaction details in interactive information services in a telephonic communication system H04M 3/4936)}

G10L 15/28 ・Constructional details of speech recognition systems

G10L 15/285 ・・{Memory allocation or algorithm optimisation to reduce hardware requirements}

G10L 15/30 ・・Distributed recognition, e.g. in client-server systems, for mobile phones or network applications

G10L 15/32 ・・Multiple recognisers used in sequence or in parallel; Score combination systems therefor, e.g. voting systems

G10L 15/34 ・・Adaptation of a single recogniser for parallel processing, e.g. by use of multiple processors or cloud computing

G10L 17/00 Speaker identification or verification

G10L 17/005 ・{Speaker recognisers specially adapted for particular applications (G07C 9/00071 takes precedence)}

G10L 17/02 ・Preprocessing operations, e.g. segment selection; Pattern representation or modelling, e.g. based on linear discriminant analysis [LDA] or principal components; Feature selection or extraction

G10L 17/04 ・Training, enrolment or model building

G10L 17/06 ・Decision making techniques; Pattern matching strategies

G10L 17/08 ・・Use of distortion metrics or a particular distance between probe pattern and reference templates

G10L 17/10 ・・Multimodal systems, i.e. based on the integration of multiple recognition engines or fusion of expert systems

G10L 17/12 ・・Score normalisation

G10L 17/14 ・・Use of phonemic categorisation or speech recognition prior to speaker recognition or verification

G10L 17/16 ・Hidden Markov models [HMMs]

G10L 17/18 ・Artificial neural networks; Connectionist approaches

G10L 17/20 ・Pattern transformations or operations aimed at increasing system robustness, e.g. against channel noise or different working conditions

G10L 17/22 ・Interactive procedures; Man-machine interfaces

G10L 17/24 ・・the user being prompted to utter a password or a predefined phrase

G10L 17/26 ・Recognition of special voice characteristics, e.g. for use in lie detectors; Recognition of animal voices

G10L 19/00 Speech or audio signal analysis-synthesis techniques for redundancy reduction, e.g. in vocoders; Coding or decoding of speech or audio signal, using source filter models or psychoacoustic analysis (in musical instruments G10H)

G10L 19/0017 ・{Lossless audio signal coding; Perfect reconstruction of coded audio signal by transmission of coding error (G10L 19/24 takes precedence)}

G10L 19/0018 ・{Speech coding using phonetic or linguistical decoding of the source; Reconstruction using text-to-speech synthesis}

G10L 19/0019 ・{Vocoders specially adapted for particular applications}

G10L 19/002 ・Dynamic bit allocation (for perceptual audio coders G10L 19/032)

G10L 19/005 ・Correction of errors induced by the transmission channel, if related to the coding algorithm

G10L 19/008 ・Multichannel audio signal coding or decoding, i.e. using interchannel correlation to reduce redundancies, e.g. joint-stereo, intensity-coding, matrixing (arrangements for reproducing spatial sound H04R 5/00; stereophonic systems, e.g. spatial sound capture or matrixing of audio signals in the decoded state H04S)

G10L 19/012 ・Comfort noise or silence coding

G10L 19/018 ・Audio watermarking, i.e. embedding inaudible data in the audio signal

G10L 19/02 ・using spectral analysis, e.g. transform vocoders or subband vocoders

G10L 19/0204 ・・{using subband decomposition}

G10L 19/0208 ・・・{Subband vocoders}

G10L 19/0212 ・・{using orthogonal transformation}

G10L 19/0216 ・・・{using wavelet decomposition}

G10L 19/022 ・・Blocking, i.e. grouping of samples in time; Choice of analysis windows; Overlap factoring

G10L 19/025 ・・・Detection of transients or attacks for time/frequency resolution switching

G10L 19/028 ・・Noise substitution, i.e. substituting non-tonal spectral components by noisy source (comfort noise for discontinuous speech transmission G10L 19/012)

G10L 19/03 ・・Spectral prediction for preventing pre-echo; Temporary noise shaping [TNS], e.g. in MPEG2 or MPEG4

G10L 19/032 ・・Quantisation or dequantisation of spectral components

G10L 19/035 ・・・Scalar quantisation

G10L 19/038 ・・・Vector quantisation, e.g. TwinVQ audio

G10L 19/04 ・using predictive techniques

G10L 19/06 ・・Determination or coding of the spectral characteristics, e.g. of the short-term prediction coefficients

G10L 19/07 ・・・Line spectrum pair [LSP] vocoders

G10L 19/08 ・・Determination or coding of the excitation function ; Determination or coding of the long-term prediction parameters

G10L 19/083 ・・・the excitation function being an excitation gain (G10L 25/90 takes precedence)

G10L 19/087 ・・・using mixed excitation models, e.g. MELP, MBE, split band LPC or HVXC

G10L 19/09 ・・・Long term prediction, i.e. removing periodical redundancies, e.g. by using adaptive codebook or pitch predictor

G10L 19/093 ・・・using sinusoidal excitation models

G10L 19/097 ・・・using prototype waveform decomposition or prototype waveform interpolative [PWI] coders

G10L 19/10 ・・・the excitation function being a multipulse excitation

G10L 19/107 ・・・・Sparse pulse excitation, e.g. by using algebraic codebook

G10L 19/113 ・・・・Regular pulse excitation

G10L 19/12 ・・・the excitation function being a code excitation, e.g. in code excited linear prediction [CELP] vocoders

G10L 19/125 ・・・・Pitch excitation, e.g. pitch synchronous innovation CELP [PSI-CELP]

G10L 19/13 ・・・・Residual excited linear prediction [RELP]

G10L 19/135 ・・・・Vector sum excited linear prediction [VSELP]

G10L 19/16 ・・Vocoder architecture

G10L 19/167 ・・・{Audio streaming, i.e. formatting and decoding of an encoded audio signal representation into a data stream for transmission or storage purposes}

G10L 19/173 ・・・{Transcoding, i.e. converting between two coded representations avoiding cascaded coding-decoding}

G10L 19/18 ・・・Vocoders using multiple modes

G10L 19/20 ・・・・using sound class specific coding, hybrid encoders or object based coding

G10L 19/22 ・・・・Mode decision, i.e. based on audio signal content versus external parameters

G10L 19/24 ・・・・Variable rate codecs, e.g. for generating different qualities using a scalable representation such as hierarchical encoding or layered encoding

G10L 19/26 ・・Pre-filtering or post-filtering

G10L 19/265 ・・・{Pre-filtering, e.g. high frequency emphasis prior to encoding}

G10L 21/00 Processing of the speech or voice signal to produce another audible or non-audible signal, e.g. visual or tactile, in order to modify its quality or its intelligibility (G10L 19/00 takes precedence)

G10L 21/003 ・Changing voice quality, e.g. pitch or formants

G10L 21/007 ・・characterised by the process used

G10L 21/01 ・・・Correction of time axis

G10L 21/013 ・・・Adapting to target pitch

G10L 21/02 ・Speech enhancement, e.g. noise reduction or echo cancellation (reducing echo effects in line transmission systems H04B 3/20 ; echo suppression in hands-free telephones H04M 9/08)

G10L 21/0202 ・・{Applications}

G10L 21/0205 ・・・{Enhancement of intelligibility of clean or coded speech}

G10L 21/0208 ・・Noise filtering

G10L 21/0216 ・・・characterised by the method used for estimating noise

G10L 21/0224 ・・・・Processing in the time domain

G10L 21/0232 ・・・・Processing in the frequency domain

G10L 21/0264 ・・・characterised by the type of parameter measurement, e.g. correlation techniques, zero crossing techniques or predictive techniques

G10L 21/0272 ・・Voice signal separating

G10L 21/028 ・・・using properties of sound source

G10L 21/0308 ・・・characterised by the type of parameter measurement, e.g. correlation techniques, zero crossing techniques or predictive techniques

G10L 21/0316 ・・by changing the amplitude

G10L 21/0324 ・・・Details of processing therefor

G10L 21/0332 ・・・・involving modification of waveforms

G10L 21/034 ・・・・Automatic adjustment

G10L 21/0356 ・・・for synchronising with other signals, e.g. video signals

G10L 21/0364 ・・・for improving intelligibility

G10L 21/038 ・・using band spreading techniques

G10L 21/0388 ・・・Details of processing therefor

G10L 21/04 ・Time compression or expansion

G10L 21/043 ・・by changing speed

G10L 21/045 ・・・using thinning out or insertion of a waveform

G10L 21/047 ・・・・characterised by the type of waveform to be thinned out or inserted

G10L 21/049 ・・・・characterised by the interconnection of waveforms

G10L 21/055 ・・for synchronising with other signals, e.g. video signals

G10L 21/057 ・・for improving intelligibility

G10L 21/06 ・Transformation of speech into a non-audible representation, e.g. speech visualisation or speech processing for tactile aids (G10L 15/26 takes precedence)

G10L 21/10 ・・transforming into visible information

G10L 21/12 ・・・by displaying time domain information

G10L 21/14 ・・・by displaying frequency domain information

G10L 21/16 ・・transforming into a non-visible representation (devices or methods enabling ear patients to replace direct auditory perception by another kind of perception A61F 11/04)

G10L 21/18 ・・Details of the transformation process

G10L 25/00 Speech or voice analysis techniques not restricted to a single one of groups G10L 15/00-G10L 21/00

G10L 25/03 ・characterised by the type of extracted parameters

G10L 25/06 ・・the extracted parameters being correlation coefficients

G10L 25/09 ・・the extracted parameters being zero crossing rates

G10L 25/12 ・・the extracted parameters being prediction coefficients

G10L 25/15 ・・the extracted parameters being formant information

G10L 25/18 ・・the extracted parameters being spectral information of each sub-band

G10L 25/21 ・・the extracted parameters being power information

G10L 25/24 ・・the extracted parameters being the cepstrum

G10L 25/27 ・characterised by the analysis technique

G10L 25/30 ・・using neural networks

G10L 25/33 ・・using fuzzy logic

G10L 25/36 ・・using chaos theory

G10L 25/39 ・・using genetic algorithms

G10L 25/45 ・characterised by the type of analysis window

G10L 25/48 ・specially adapted for particular use

G10L 25/51 ・・for comparison or discrimination

G10L 25/54 ・・・for retrieval

G10L 25/57 ・・・for processing of video signals

G10L 25/60 ・・・for measuring the quality of voice signals

G10L 25/63 ・・・for estimating an emotional state

G10L 25/66 ・・・for extracting parameters related to health condition (detecting or measuring for diagnostic purposes A61B 5/00)

G10L 25/69 ・・for evaluating synthetic or decoded voice signals

G10L 25/72 ・・for transmitting results of analysis

G10L 25/75 ・for modelling vocal tract parameters

G10L 25/78 ・Detection of presence or absence of voice signals (switching of direction of transmission by voice frequency in two-way loud-speaking telephone systems H04M 9/10)

G10L 25/81 ・・for discriminating voice from music

G10L 25/84 ・・for discriminating voice from noise

G10L 25/87 ・・Detection of discrete points within a voice signal

G10L 25/90 ・Pitch determination of speech signals

G10L 25/93 ・Discriminating between voiced and unvoiced parts of speech signals (G10L 25/90 takes precedence)

G10L 99/00 Subject matter not provided for in other groups of this subclass

G10L 2013/00 Speech synthesis; Text to speech systems

G10L 2013/02 ・Methods for producing synthetic speech; Speech synthesisers

G10L 2013/021 ・・{Overlap-add techniques}

G10L 2013/08 ・Text analysis or generation of parameters for speech synthesis out of text, e.g. grapheme to phoneme translation, prosody generation or stress or intonation determination

G10L 2013/083 ・・{Special characters, e.g. punctuation marks}

G10L 2013/10 ・・Prosody rules derived from text; Stress or intonation

G10L 2013/105 ・・・{Duration}

G10L 2015/00 Speech recognition (G10L 17/00 takes precedence)

G10L 2015/02 ・Feature extraction for speech recognition; Selection of recognition unit

G10L 2015/022 ・・{Demisyllables, biphones or triphones being the recognition units}

G10L 2015/025 ・・{Phonemes, fenemes or fenones being the recognition units}

G10L 2015/027 ・・{Syllables being the recognition units}

G10L 2015/06 ・Creation of reference templates ; Training of speech recognition systems, e.g. adaptation to the characteristics of the speaker's voice (G10L 15/14 takes precedence)

G10L 2015/063 ・・{Training}

G10L 2015/0631 ・・・{Creating reference templates; Clustering}

G10L 2015/0633 ・・・・{using lexical or orthographic knowledge sources}

G10L 2015/0635 ・・・{updating or merging of old and new templates; Mean values; Weighting}

G10L 2015/0636 ・・・・{Threshold criteria for the updating}

G10L 2015/0638 ・・・{Interactive procedures}

G10L 2015/08 ・Speech classification or search

G10L 2015/081 ・・{Search algorithms, e.g. Baum-Welch or Viterbi}

G10L 2015/085 ・・{Methods for reducing search complexity, pruning}

G10L 2015/086 ・・{Recognition of spelled words}

G10L 2015/088 ・・{Word spotting}

G10L 2015/22 ・Procedures used during a speech recognition process, e.g. man-machine dialogue

G10L 2015/221 ・・Announcement of recognition results

G10L 2015/223 ・・Execution procedure of a spoken command

G10L 2015/225 ・・Feedback of the input speech

G10L 2015/226 ・・Taking into account non-speech caracteristics

G10L 2015/227 ・・・of the speaker; Human-factor methodology

G10L 2015/228 ・・・of application context

G10L 2019/00 Speech or audio signal analysis-synthesis techniques for redundancy reduction, e.g. in vocoders; Coding or decoding of speech or audio signal, using source filter models or psychoacoustic analysis (in musical instruments G10H)

G10L 2019/0001 ・{Codebooks }

G10L 2019/0002 ・・{Codebook adaptations}

G10L 2019/0003 ・・{Backward prediction of gain}

G10L 2019/0004 ・・{Design or structure of the codebook}

G10L 2019/0005 ・・・{Multi-stage vector quantisation}

G10L 2019/0006 ・・・{Tree or treillis structures; Delayed decisions}

G10L 2019/0007 ・・{Codebook element generation}

G10L 2019/0008 ・・・{Algebraic codebooks}

G10L 2019/0009 ・・・{Orthogonal codebooks}

G10L 2019/001 ・・・{Interpolation of codebook vectors}

G10L 2019/0011 ・・{Long term prediction filters, i.e. pitch estimation}

G10L 2019/0012 ・・{Smoothing of parameters of the decoder interpolation}

G10L 2019/0013 ・・{Codebook search algorithms}

G10L 2019/0014 ・・・{Selection criteria for distances}

G10L 2019/0015 ・・・{Viterbi algorithms}

G10L 2019/0016 ・・{Codebook for LPC parameters}

G10L 2021/00 Processing of the speech or voice signal to produce another audible or non-audible signal, e.g. visual or tactile, in order to modify its quality or its intelligibility (G10L 19/00 takes precedence)

G10L 2021/003 ・Changing voice quality, e.g. pitch or formants

G10L 2021/007 ・・characterised by the process used

G10L 2021/013 ・・・Adapting to target pitch

G10L 2021/0135 ・・・・{Voice conversion or morphing}

G10L 2021/02 ・Speech enhancement, e.g. noise reduction or echo cancellation (reducing echo effects in line transmission systems H04B 3/20 ; echo suppression in hands-free telephones H04M 9/08)

G10L 2021/0208 ・・Noise filtering

G10L 2021/02082 ・・・{the noise being echo, reverberation of the speech}

G10L 2021/02085 ・・・{Periodic noise}

G10L 2021/02087 ・・・{the noise being separate speech, e.g. cocktail party}

G10L 2021/0216 ・・・characterised by the method used for estimating noise

G10L 2021/02161 ・・・・{Number of inputs available containing the signal or the noise to be suppressed}

G10L 2021/02163 ・・・・・{Only one microphone}

G10L 2021/02165 ・・・・・[N: Two microphones, one receiving mainly the noise signal and the other one mainly the speech signal

G10L 2021/02166 ・・・・・{Microphone arrays; Beamforming}

G10L 2021/02168 ・・・・{the estimation exclusively taking place during speech pauses}

G10L 2021/0316 ・・by changing the amplitude

G10L 2021/0364 ・・・for improving intelligibility

G10L 2021/03643 ・・・・{Diver speech}

G10L 2021/03646 ・・・・{Stress or Lombard effect}

G10L 2021/04 ・Time compression or expansion

G10L 2021/057 ・・for improving intelligibility

G10L 2021/0575 ・・・{Aids for the handicapped in speaking}

G10L 2021/06 ・Transformation of speech into a non-audible representation, e.g. speech visualisation or speech processing for tactile aids (G10L 15/26 takes precedence)

G10L 2021/065 ・・{Aids for the handicapped in understanding}

G10L 2021/10 ・・transforming into visible information

G10L 2021/105 ・・・{Synthesis of the lips movements from speech, e.g. for talking heads}

G10L 2025/00 Speech or voice analysis techniques not restricted to a single one of groups G10L 15/00-G10L 21/00

G10L 2025/78 ・Detection of presence or absence of voice signals (switching of direction of transmission by voice frequency in two-way loud-speaking telephone systems H04M 9/10)

G10L 2025/783 ・・{based on threshold decision}

G10L 2025/786 ・・・{Adaptive threshold}

G10L 2025/90 ・Pitch determination of speech signals

G10L 2025/903 ・・{using a laryngograph}

G10L 2025/906 ・・{Pitch tracking}

G10L 2025/93 ・Discriminating between voiced and unvoiced parts of speech signals (G10L 25/90 takes precedence)

G10L 2025/932 ・・{Decision in previous or following frames}

G10L 2025/935 ・・{Mixed voiced class; Transitions}

G10L 2025/937 ・・{Signal energy in various frequency bands}