| Patents > Guidance, Tools, and Manuals > Classification > Class Schedule |
|
|
|
| Class Numbers & Titles | Class Numbers Only | USPC Index | International | HELP |
|
| You are viewing a USPC Schedule. |
|
| Class 704 | DATA PROCESSING: SPEECH SIGNAL PROCESSING, LINGUISTICS, LANGUAGE TRANSLATION, AND AUDIO COMPRESSION/DECOMPRESSION |
| Click here for a printable version of this file | |
![[List of Pre Grant Publications for class 704 subclass 1]](../as.gif) | ![[List of Patents for class 704 subclass 1]](../ps.gif) | 1 | LINGUISTICS |
| | 2 |  Translation machine |
| | 3 |  Having particular Input/Output device |
| | 4 | Based on phrase, clause, or idiom |
| | 5 | For partial translation |
| | 6 | Punctuation |
| | 7 | Storage or retrieval of data |
| | 8 | Multilingual or national language support |
| | 9 | Natural language |
| | 10 | Dictionary building, modification, or prioritization |
| | 200 | SPEECH SIGNAL PROCESSING |
| | 200.1 | Psychoacoustic |
| | 201 | For storage or transmission |
| | 202 | Neural network |
| | 203 | Transformation |
| | 205 | Frequency |
| | 211 | Time |
| | 212 |  Pulse code modulation (PCM) |
| | 213 | Zero crossing |
| | 214 | Voiced or unvoiced |
| | 215 | Silence decision |
| | 216 | Correlation function |
| | 219 | Linear prediction |
| | 220 | Analysis by synthesis |
| | 221 | Pattern matching vocoders |
| | 224 | Normalizing |
| | 225 | Gain control |
| | 226 | Noise |
| | 229 | Adaptive bit allocation |
| | 230 | Quantization |
| | 231 | Recognition |
| | 232 | Neural network |
| | 233 | Detect speech in noise |
| | 234 | Normalizing |
| | 235 | Speech to image |
| | 236 | Specialized equations or comparisons |
| | 237 | Correlation |
| | 238 | Distance |
| | 239 | Similarity |
| | 240 | Probability |
| | 241 | Dynamic time warping |
| | 242 | Viterbi trellis |
| | 243 | Creating patterns for matching |
| | 246 | Voice recognition |
| | 251 | Word recognition |
| | 252 | Preliminary matching |
| | 253 | Endpoint detection |
| | 254 | Subportions |
| | 255 | Specialized models |
| | 256 |  Markov |
| | 256.1 |  Hidden Markov Model (HMM) (EPO) |
| | 256.2 |  Training of HMM (EPO) |
| | 256.3 |  With insufficient amount of training data, e.g., state sharing, tying, deleted interpolation (EPO) |
| | 256.4 | Duration modeling in HMM, e.g., semi HMM, segmental models, transition probabilities (EPO) |
| | 256.5 | Hidden Markov (HM) network (EPO) |
| | 256.6 | State emission probability (EPO) |
| | 257 | Natural language |
| | 258 | Synthesis |
| | 259 | Neural network |
| | 260 | Image to speech |
| | 261 | Vocal tract model |
| | 262 | Linear prediction |
| | 263 | Correlation |
| | 264 | Excitation |
| | 265 | Interpolation |
| | 266 | Specialized model |
| | 267 | Time element |
| | 268 | Frequency element |
| | 269 | Transformation |
| | 270 | Application |
| | 500 | AUDIO SIGNAL BANDWIDTH COMPRESSION OR EXPANSION |
| | 503 | AUDIO SIGNAL TIME COMPRESSION OR EXPANSION (E.G., RUN LENGTH CODING) |
| E-SUBCLASSES | ||
| The following subclasses beginning with the letter E are E-subclasses. Each E-subclass corresponds in scope to a classification in a foreign classification system, for example, the European Classification system (ECLA). The foreign classification equivalent to an E-subclass is identified in the subclass definition. In addition to US documents classified in E-subclasses by US examiners, documents are regularly classified in E-subclasses according to the classification practices of any foreign Offices identified in parentheses at the end of the title. For example, "(EPO)" at the end of a title indicates both European and US patent documents, as classified by the EPO, are regularly added to the subclass. E-subclasses may contain subject matter outside the scope of this class.Consult their definitions, or the documents themselves to clarify or interpret titles. |
| | E17.001 | SPEAKER IDENTIFICATION OR VERIFICATION (EPO) |
| | E17.002 | Recognition of special voice characteristics, e.g., for use in a lie detector; recognition of animal voices, etc. (EPO) |
| | E17.003 | Systems using speaker recognizers (EPO) |
| | E17.004 | Details (EPO) |
| | E17.006 | Training, model building, enrollment (EPO) |
| | E17.007 | Decision making techniques, pattern matching strategies (EPO) |
| | E17.008 | Use of particular distance or distortion metric between probe pattern and reference templates (EPO) |
| | E17.009 | Multimodal systems, i.e., based on the integration of multiple recognition engines or experts fusion (EPO) |
| | E17.01 | Score normalization (EPO) |
| | E17.011 | Use of phonemic categorization or speech recognition prior to speaker recognition or verification (EPO) |
| | E17.012 | Hidden Markov Models (HMMs) (EPO) |
| | E17.013 | Artificial neural networks, connectionist approaches (EPO) |
| | E17.014 | Pattern transformations and operations aimed at increasing system robustness, e.g., against channel noise, different working conditions, etc. (EPO) |
| | E17.015 | Interactive procedures, man-machine interface (EPO) |
| |