CN108711419B - Environmental sound sensing method and system for cochlear implant - Google Patents
Environmental sound sensing method and system for cochlear implant Download PDFInfo
- Publication number
- CN108711419B CN108711419B CN201810856692.8A CN201810856692A CN108711419B CN 108711419 B CN108711419 B CN 108711419B CN 201810856692 A CN201810856692 A CN 201810856692A CN 108711419 B CN108711419 B CN 108711419B
- Authority
- CN
- China
- Prior art keywords
- module
- sound
- neural network
- characteristic values
- group
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 27
- 230000007613 environmental effect Effects 0.000 title claims abstract description 19
- 239000007943 implant Substances 0.000 title claims abstract description 18
- 238000013528 artificial neural network Methods 0.000 claims abstract description 67
- 238000012545 processing Methods 0.000 claims abstract description 27
- 230000005236 sound signal Effects 0.000 claims abstract description 27
- 238000000605 extraction Methods 0.000 claims abstract description 26
- 239000000284 extract Substances 0.000 claims abstract description 7
- 230000008569 process Effects 0.000 claims abstract description 6
- 238000004422 calculation algorithm Methods 0.000 claims description 14
- 238000012549 training Methods 0.000 claims description 12
- 210000002569 neuron Anatomy 0.000 claims description 6
- 230000004913 activation Effects 0.000 claims description 5
- 239000011159 matrix material Substances 0.000 claims description 4
- 230000003111 delayed effect Effects 0.000 claims description 3
- 238000011156 evaluation Methods 0.000 claims description 3
- 239000000203 mixture Substances 0.000 claims description 3
- 238000005070 sampling Methods 0.000 claims description 3
- 238000012216 screening Methods 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 2
- 230000006870 function Effects 0.000 description 5
- 230000008447 perception Effects 0.000 description 5
- 210000003477 cochlea Anatomy 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000000638 stimulation Effects 0.000 description 2
- 238000012706 support-vector machine Methods 0.000 description 2
- 206010011878 Deafness Diseases 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 238000007635 classification algorithm Methods 0.000 description 1
- 210000000860 cochlear nerve Anatomy 0.000 description 1
- 231100000895 deafness Toxicity 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000005484 gravity Effects 0.000 description 1
- 208000016354 hearing loss disease Diseases 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04R—LOUDSPEAKERS, MICROPHONES, GRAMOPHONE PICK-UPS OR LIKE ACOUSTIC ELECTROMECHANICAL TRANSDUCERS; DEAF-AID SETS; PUBLIC ADDRESS SYSTEMS
- H04R25/00—Deaf-aid sets, i.e. electro-acoustic or electro-mechanical hearing aids; Electric tinnitus maskers providing an auditory perception
- H04R25/50—Customised settings for obtaining desired overall acoustical characteristics
- H04R25/505—Customised settings for obtaining desired overall acoustical characteristics using digital signal processing
- H04R25/507—Customised settings for obtaining desired overall acoustical characteristics using digital signal processing implemented by neural network or fuzzy logic
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/02—Feature extraction for speech recognition; Selection of recognition unit
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/06—Creation of reference templates; Training of speech recognition systems, e.g. adaptation to the characteristics of the speaker's voice
- G10L15/063—Training
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/08—Speech classification or search
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L21/00—Speech or voice signal processing techniques to produce another audible or non-audible signal, e.g. visual or tactile, in order to modify its quality or its intelligibility
- G10L21/02—Speech enhancement, e.g. noise reduction or echo cancellation
- G10L21/0208—Noise filtering
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
- G10L25/27—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the analysis technique
- G10L25/30—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the analysis technique using neural networks
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
- G10L25/48—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use
- G10L25/51—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use for comparison or discrimination
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04R—LOUDSPEAKERS, MICROPHONES, GRAMOPHONE PICK-UPS OR LIKE ACOUSTIC ELECTROMECHANICAL TRANSDUCERS; DEAF-AID SETS; PUBLIC ADDRESS SYSTEMS
- H04R25/00—Deaf-aid sets, i.e. electro-acoustic or electro-mechanical hearing aids; Electric tinnitus maskers providing an auditory perception
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Human Computer Interaction (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Computational Linguistics (AREA)
- Signal Processing (AREA)
- Artificial Intelligence (AREA)
- Neurosurgery (AREA)
- Otolaryngology (AREA)
- General Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Fuzzy Systems (AREA)
- Automation & Control Theory (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Quality & Reliability (AREA)
- Circuit For Audible Band Transducer (AREA)
- Prostheses (AREA)
- Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
Abstract
The invention discloses an environment sound sensing method and system of a cochlear implant, wherein the method comprises the following steps: the sound collection module collects environmental sound in real time by adopting a microphone and outputs a section of collected discrete sound signal to the sound characteristic extraction module; the sound characteristic extraction module processes the sound signals sent by the sound acquisition module, extracts a group of characteristic values representing the characteristics of the sound signals and outputs the characteristic values to the neural network classification module; the neural network classification module receives a group of characteristic values extracted by the sound characteristic extraction module, classifies the group of characteristic values through the trained neural network, and then outputs the classification result to the comprehensive decision module; after receiving the classification result of the neural network classification module, the comprehensive decision module comprehensively analyzes and gives judgment of the current scene, and outputs the judgment result to the voice processing selection module; and the voice processing selection module selects an optimal voice processing program and parameter configuration thereof according to the judgment result of the comprehensive decision module on the current scene.
Description
Technical Field
The invention belongs to the field of signal processing, and relates to an environment sound sensing method and system for a cochlear implant.
Background
The artificial cochlea is the only medical instrument which can effectively recover the hearing of the patients with severe or extremely severe deafness in the current market. The general working principle of the cochlear implant is that a sound signal collected by a microphone is converted into a stimulation code through a signal processing unit and is sent to an implant, and the implant stimulates an auditory nerve through a microelectrode according to the stimulation code, so that an implant recovers the hearing. Like other hearing aids, such systems lack an important function of the hearing system of a regular person, i.e. the target signal can be distinguished and extracted in a complex sound scene. Such as listening to words spoken by chat objects in a group of people or in a relatively noisy environment. The common solution is to reduce the influence of noise on the hearing through a certain denoising algorithm. However, the denoising algorithm and the parameter configuration of the algorithm are different under different environments (such as pure voice, voice to be denoised or noise environments).
In order to solve the problems, an ambient sound perception algorithm is introduced, and the system can pertinently start the noise reduction algorithm and configure related parameters according to the judgment result of the ambient sound perception algorithm. In early cochlear implant or hearing aid systems, hidden markov models were used as classifiers for the ambient sound perception algorithm. The model is relatively simple, the theory is mature earlier, the requirement on training data is not high, and a certain correct recognition rate is also kept. And the operation amount is low, and the method can be suitable for the device with limited operation capability, such as the cochlear implant. With continuous innovation in the fields of pattern recognition, machine learning and the like and continuous progress in computational algorithm in recent years, more classification algorithms (support vector machines, neural networks and the like) are more prominent in the field of environmental sound perception, and the classification accuracy is higher. And classifiers such as support vector machines, neural networks, and the like place the center of gravity on the distinguishing categories without providing prior probability of category transformation, relative to the hidden Markov model. That is, only data of different ambient sounds need to be analyzed, without considering what the probability of converting one ambient sound to another. It is very difficult to obtain such transition probabilities and the analysis from the data is not accurate enough. However, the neural network varies greatly, and the network structure can be combined in many ways according to the number of input characteristic values, the number of hidden layers and the number of nodes in each layer. In addition, the classification accuracy of the neural network is generally proportional to the scale of the neural network, and thus the required computation amount is relatively large.
Disclosure of Invention
In order to solve the problems, the invention provides an environmental sound sensing method of an artificial cochlea aiming at the defects of the existing sound sensing processing, a neural network is adopted to classify the environmental sound, the input characteristic value and the network structure of the neural network are optimized on an artificial cochlea system, and the calculation amount is minimized under the condition of meeting a certain classification accuracy.
In order to achieve the above object, the technical solution of the present invention is an ambient sound sensing method for a cochlear implant, including the steps of:
the sound collection module collects environmental sound in real time by adopting a microphone and then outputs a section of collected discrete sound signal to the sound feature extraction module;
the sound characteristic extraction module processes the sound signals sent by the sound acquisition module, extracts a group of characteristic values representing the characteristics of the sound signals and outputs the characteristic values to the neural network classification module;
after receiving a group of characteristic values extracted by the sound characteristic extraction module, the neural network classification module classifies the group of characteristic values through the trained neural network and then outputs the classification result to the comprehensive decision-making module;
after receiving the classification result of the neural network classification module, the comprehensive decision module comprehensively analyzes and gives judgment of the current scene, and outputs the judgment result to the voice processing selection module;
and the voice processing selection module selects an optimal voice processing program and parameter configuration thereof according to the judgment result of the comprehensive decision module on the current scene.
Preferably, the microphones collect ambient sound in real time using an omni-directional microphone or a microphone array.
Preferably, the sampling rate of the sound collection module is 16K.
Preferably, the number of feature values extracted from the set of feature values representing the characteristics of the sound signal is 8.
Preferably, the neural network classification module employs a deep neural network or a delayed neural network comprising two hidden layers, 15 neurons per layer.
Preferably, 8 of said characteristic values are selected from 60 characteristic values.
Preferably, the feature value screening adopts a method of comprehensively analyzing statistics of feature values and a gaussian mixture model, an average influence value algorithm, a sequence forward selection algorithm and classifier training result evaluation.
Preferably, the calculation amount of the characteristic value and the calculation amount of the neural network do not exceed 20% of the operation capacity of the artificial cochlear speech processor.
Based on the above purpose, the invention also provides an environmental sound perception system of the cochlear implant, which comprises a sound collection module, a sound feature extraction module, a neural network classification module, a comprehensive decision module and a voice processing selection module which are connected in sequence, wherein,
the sound collection module is used for collecting environmental sound in real time by adopting a microphone and then outputting a section of collected discrete sound signal to the sound feature extraction module;
the sound feature extraction module is used for processing the sound signals sent by the sound acquisition module, extracting a group of feature values representing the characteristics of the sound signals and outputting the feature values to the neural network classification module;
the neural network classification module is used for classifying a group of characteristic values extracted by the sound characteristic extraction module through the trained neural network after receiving the group of characteristic values, and then outputting the classification result to the comprehensive decision module;
the comprehensive decision-making module is used for comprehensively analyzing and giving judgment of the current scene after receiving the classification result of the neural network classification module and outputting the judgment result to the voice processing selection module;
and the voice processing selection module is used for selecting an optimal voice processing program and parameter configuration thereof according to the judgment result of the comprehensive decision module on the current scene.
Drawings
Fig. 1 is a flowchart illustrating steps of a cochlear implant ambient sound sensing method according to an embodiment of the present invention;
fig. 2 is a block diagram of an ambient sound sensing system of a cochlear implant according to an embodiment of the present invention;
fig. 3 is a detailed schematic diagram of a neural network classification module of the cochlear implant environmental sound sensing method and system according to the embodiment of the present invention;
fig. 4 is a comparison graph of the operation amount and the accuracy of the network with different hidden layers and different neuron numbers according to the method for sensing the environmental sound of the cochlear implant of the embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
On the contrary, the invention is intended to cover alternatives, modifications, equivalents and alternatives which may be included within the spirit and scope of the invention as defined by the appended claims. Furthermore, in the following detailed description of the present invention, certain specific details are set forth in order to provide a better understanding of the present invention. It will be apparent to one skilled in the art that the present invention may be practiced without these specific details.
Referring to fig. 1, a flow chart of steps of the method for sensing environmental sound of cochlear implant according to the technical solution of the embodiment of the present invention includes the following steps:
s10, the sound collection module collects the environment sound in real time by a microphone and then outputs a section of collected discrete sound signal to the sound feature extraction module;
s20, the sound feature extraction module processes the sound signal sent by the sound collection module, extracts a group of feature values representing the characteristics of the sound signal and outputs the feature values to the neural network classification module;
s30, after receiving a group of characteristic values extracted by the sound characteristic extraction module, the neural network classification module classifies the group of characteristic values through the trained neural network, and then outputs the classification result to the comprehensive decision module;
s40, after receiving the classification result of the neural network classification module, the comprehensive decision module comprehensively analyzes and gives the judgment of the current scene, and outputs the judgment result to the voice processing selection module;
and S50, selecting the optimal voice processing program and the parameter configuration thereof by the voice processing selection module according to the judgment result of the comprehensive decision module on the current scene.
Referring to fig. 2, the system of the present invention includes a sound collection module 10, a sound feature extraction module 20, a neural network classification module 30, a comprehensive decision module 40, and a speech processing selection module 50, which are connected in sequence, wherein,
the sound collection module 10 is configured to collect environmental sound in real time by using a microphone, and then output a section of collected discrete sound signals to the sound feature extraction module 20;
the sound feature extraction module 20 is configured to process the sound signal sent by the sound acquisition module, extract a group of feature values representing characteristics of the sound signal, and output the feature values to the neural network classification module 30;
the neural network classification module 30 is configured to, after receiving a set of feature values extracted by the sound feature extraction module, classify the set of feature values through the trained neural network, and then output a classification result to the comprehensive decision module 40;
the comprehensive decision module 40 is configured to, after receiving the classification result of the neural network classification module, comprehensively analyze the classification result to give a judgment of the current scene, and output the judgment result to the voice processing selection module 50;
and the voice processing selection module 50 is used for selecting an optimal voice processing program and parameter configuration thereof according to the judgment result of the comprehensive decision module on the current scene.
In a specific embodiment, in S10, the microphone collects the ambient sound in real time by using an omnidirectional microphone or a microphone array, and the sampling rate of the sound collection module 10 is 16K.
In S20, a group of feature values representing characteristics of the sound signal are extracted, and 8 feature values are extracted from 60 feature values. Before extracting the characteristic value, normalization processing is carried out, and the formula is as follows:
wherein x isnormTo normalize the result, XmaxFor the maximum value of the training sample, X, where the characteristic value is locatedminThe minimum value of the training sample where the characteristic value is located.
The neural network classification module in S30 adopts a deep neural network or a delayed neural network including two hidden layers, 15 neurons per layer. The neural network module is obtained by training a large number of data samples, taking 4 types of environmental sounds (pure voice, noisy voice, noise, music and silence) as an example, and the neural network model thereof is shown in fig. 3. The characteristic values are selected from 1, 2, 3, 4, 5 and 6, and six types are combined into a group. The training samples are extracted from a large number of collected audio files, and comprise 144000 groups of sample characteristic values in total, and each type of environmental sound comprises 36000 groups of characteristic values. To find a balance between the amount of computation and the accuracy, we tried 1 hidden layer and 2 hidden layer, each with different numbers of neurons, see fig. 4. As can be seen from the figure, the accuracy of the neural network of the two hidden layers is obviously higher than that of the neural network of the single hidden layer, and the optimal neuron number is 15.
The neural network decision formula in S40 is as follows:
wherein, XinputFor inputting a matrix of eigenvalues, W1、W2、W3For each layer of the trained neural network, B1、B2、B3For each layer of bias matrix of the trained neural network, activeFcn is an activation function, YoutThe results are calculated for the network.
To reduce the amount of computation, we will use the activation function activeFcn of the hidden layerHAnd an activation function activeFcn of the output layerOAre respectively defined as:
where x is the input to the activation function and i is the ambient sound category.
After receiving the classification result of the neural network classification module, the comprehensive decision module comprehensively analyzes a series of factors, mainly including the recognition result of the neural network and the sound energy in a short period of time, gives the judgment of the current scene, and outputs the judgment result to the voice processing selection module.
And the voice processing selection module selects an optimal voice processing program and parameter configuration thereof according to the judgment result of the comprehensive decision module on the current scene.
The characteristic value screening adopts a method of comprehensively analyzing the statistic value of the characteristic value, a Gaussian mixture model, an average influence value algorithm, a sequence forward selection algorithm and classifier training result evaluation.
The calculated amount of the characteristic value and the calculated amount of the neural network do not exceed 20% of the operation capacity of the artificial cochlea speech processor.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents and improvements made within the spirit and principle of the present invention are intended to be included within the scope of the present invention.
Claims (6)
1. An ambient sound sensing method of a cochlear implant, comprising the steps of:
the sound collection module collects environmental sound in real time by adopting a microphone and then outputs a section of collected discrete sound signal to the sound feature extraction module;
the sound characteristic extraction module processes the sound signals sent by the sound acquisition module, extracts a group of characteristic values representing the characteristics of the sound signals and outputs the characteristic values to the neural network classification module;
after receiving a group of characteristic values extracted by the sound characteristic extraction module, the neural network classification module classifies the group of characteristic values through the trained neural network and then outputs the classification result to the comprehensive decision-making module;
after receiving the classification result of the neural network classification module, the comprehensive decision module comprehensively analyzes and gives judgment of the current scene, and outputs the judgment result to the voice processing selection module;
the voice processing selection module selects an optimal voice processing program and parameter configuration thereof according to the judgment result of the comprehensive decision module on the current scene;
the sound characteristic extraction module processes the sound signals sent by the sound acquisition module, extracts a group of characteristic values representing the characteristics of the sound signals, outputs the characteristic values to the neural network classification module, extracts 8 characteristic values for extracting the group of characteristic values representing the characteristics of the sound signals, screens the 8 characteristic values from 60 characteristic values, and performs normalization processing before extracting the characteristic values, wherein the formula is as follows:
wherein x isnormTo normalize the result, XmaxFor the maximum value of the training sample, X, where the characteristic value is locatedminThe minimum value of the training sample where the characteristic value is located;
the neural network classification module classifies a group of characteristic values extracted by the sound characteristic extraction module through a trained neural network after receiving the group of characteristic values, then outputs the classification result to the comprehensive decision module, the neural network classification module adopts a deep neural network or a delayed neural network containing two hidden layers and 15 neurons in each layer, the neural network module is obtained through training of a large number of data samples to distinguish 4 types of environmental sounds including pure voice, noisy voice, noise, music and silence, the characteristic values are selected from 1, 2, 3, 4, 5 and 6 to form a group, the training samples are extracted from a large number of collected audio files and contain 144000 groups of sample characteristic values, and each type of environmental sound contains 36000 groups of characteristic values;
after receiving the classification result of the neural network classification module, the comprehensive decision module comprehensively analyzes and gives the judgment of the current scene, and outputs the judgment result to the voice processing selection module, wherein the judgment formula for the neural network is as follows:
wherein, XinputFor inputting a matrix of eigenvalues, W1、W2、W3For each layer of the trained neural network, B1、B2、B3For each layer of bias matrix of the trained neural network, activeFcn is an activation function, YoutThe results are calculated for the network.
2. The method of claim 1, wherein the real-time collection of ambient sound by the microphones uses an omni-directional microphone or a microphone array.
3. The method of claim 1, wherein the sound collection module has a sampling rate of 16K.
4. The method of claim 1, wherein the eigenvalue screening adopts a method of comprehensive analysis of statistics of eigenvalues and gaussian mixture model, mean influence value algorithm, sequence forward selection algorithm, and classifier training result evaluation.
5. The method of claim 1, wherein the eigenvalue and neural network are not more than 20% of the computational power of the cochlear prosthesis speech processor.
6. The system adopting the method of one of claims 1 to 5, comprising a sound collection module, a sound feature extraction module, a neural network classification module, a comprehensive decision module and a voice processing selection module which are connected in sequence, wherein,
the sound collection module is used for collecting environmental sound in real time by adopting a microphone and then outputting a section of collected discrete sound signal to the sound feature extraction module;
the sound feature extraction module is used for processing the sound signals sent by the sound acquisition module, extracting a group of feature values representing the characteristics of the sound signals and outputting the feature values to the neural network classification module;
the neural network classification module is used for classifying a group of characteristic values extracted by the sound characteristic extraction module through the trained neural network after receiving the group of characteristic values, and then outputting the classification result to the comprehensive decision module;
the comprehensive decision-making module is used for comprehensively analyzing and giving judgment of the current scene after receiving the classification result of the neural network classification module and outputting the judgment result to the voice processing selection module;
and the voice processing selection module is used for selecting an optimal voice processing program and parameter configuration thereof according to the judgment result of the comprehensive decision module on the current scene.
Priority Applications (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810856692.8A CN108711419B (en) | 2018-07-31 | 2018-07-31 | Environmental sound sensing method and system for cochlear implant |
ES202190003A ES2849124B2 (en) | 2018-07-31 | 2019-07-19 | ENVIRONMENTAL SOUND DETECTION METHOD AND SYSTEM FOR A COCHLEAR IMPLANT |
PCT/CN2019/096648 WO2020024807A1 (en) | 2018-07-31 | 2019-07-19 | Artificial cochlea ambient sound sensing method and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810856692.8A CN108711419B (en) | 2018-07-31 | 2018-07-31 | Environmental sound sensing method and system for cochlear implant |
Publications (2)
Publication Number | Publication Date |
---|---|
CN108711419A CN108711419A (en) | 2018-10-26 |
CN108711419B true CN108711419B (en) | 2020-07-31 |
Family
ID=63874461
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810856692.8A Active CN108711419B (en) | 2018-07-31 | 2018-07-31 | Environmental sound sensing method and system for cochlear implant |
Country Status (3)
Country | Link |
---|---|
CN (1) | CN108711419B (en) |
ES (1) | ES2849124B2 (en) |
WO (1) | WO2020024807A1 (en) |
Families Citing this family (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108711419B (en) * | 2018-07-31 | 2020-07-31 | 浙江诺尔康神经电子科技股份有限公司 | Environmental sound sensing method and system for cochlear implant |
CN109448703B (en) * | 2018-11-14 | 2021-05-11 | 山东师范大学 | Audio scene recognition method and system combining deep neural network and topic model |
CN111491245B (en) * | 2020-03-13 | 2022-03-04 | 天津大学 | Digital hearing aid sound field identification algorithm based on cyclic neural network and implementation method |
CN112151056B (en) * | 2020-09-27 | 2023-08-04 | 浙江诺尔康神经电子科技股份有限公司 | Intelligent cochlea sound processing system and method with customization function |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103456301A (en) * | 2012-05-28 | 2013-12-18 | 中兴通讯股份有限公司 | Ambient sound based scene recognition method and device and mobile terminal |
CN105845127A (en) * | 2015-01-13 | 2016-08-10 | 阿里巴巴集团控股有限公司 | Voice recognition method and system |
CN107103901A (en) * | 2017-04-03 | 2017-08-29 | 浙江诺尔康神经电子科技股份有限公司 | Artificial cochlea's sound scenery identifying system and method |
Family Cites Families (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1879449B (en) * | 2003-11-24 | 2011-09-28 | 唯听助听器公司 | Hearing aid and a method of noise reduction |
CN101529929B (en) * | 2006-09-05 | 2012-11-07 | Gn瑞声达A/S | A hearing aid with histogram based sound environment classification |
CN105611477B (en) * | 2015-12-27 | 2018-06-01 | 北京工业大学 | The voice enhancement algorithm that depth and range neutral net are combined in digital deaf-aid |
CN108172238B (en) * | 2018-01-06 | 2021-08-13 | 广州音书科技有限公司 | Speech enhancement algorithm based on multiple convolutional neural networks in speech recognition system |
CN108231067A (en) * | 2018-01-13 | 2018-06-29 | 福州大学 | Sound scenery recognition methods based on convolutional neural networks and random forest classification |
CN108711419B (en) * | 2018-07-31 | 2020-07-31 | 浙江诺尔康神经电子科技股份有限公司 | Environmental sound sensing method and system for cochlear implant |
-
2018
- 2018-07-31 CN CN201810856692.8A patent/CN108711419B/en active Active
-
2019
- 2019-07-19 WO PCT/CN2019/096648 patent/WO2020024807A1/en active Application Filing
- 2019-07-19 ES ES202190003A patent/ES2849124B2/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103456301A (en) * | 2012-05-28 | 2013-12-18 | 中兴通讯股份有限公司 | Ambient sound based scene recognition method and device and mobile terminal |
CN105845127A (en) * | 2015-01-13 | 2016-08-10 | 阿里巴巴集团控股有限公司 | Voice recognition method and system |
CN107103901A (en) * | 2017-04-03 | 2017-08-29 | 浙江诺尔康神经电子科技股份有限公司 | Artificial cochlea's sound scenery identifying system and method |
Non-Patent Citations (2)
Title |
---|
Speech enhancement based on neural networks improves speech intelligibility in noise for cochlear implant users;Tobias Goehring等;《Hearing Research》;20161130(第344期);第183-194页 * |
自然与城市声特征及自动识别;杨铭;《声景观 新建筑》;20141231(第5期);第32-35页 * |
Also Published As
Publication number | Publication date |
---|---|
ES2849124A1 (en) | 2021-08-13 |
WO2020024807A1 (en) | 2020-02-06 |
CN108711419A (en) | 2018-10-26 |
ES2849124B2 (en) | 2022-11-16 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108711419B (en) | Environmental sound sensing method and system for cochlear implant | |
Agrawal et al. | Novel TEO-based Gammatone features for environmental sound classification | |
CN102270451B (en) | Method and system for identifying speaker | |
CN108766459B (en) | Target speaker estimation method and system in multi-user voice mixing | |
CN108305615A (en) | A kind of object identifying method and its equipment, storage medium, terminal | |
CN107103901B (en) | Artificial cochlea sound scene recognition system and method | |
CN110197665B (en) | Voice separation and tracking method for public security criminal investigation monitoring | |
CN112151056B (en) | Intelligent cochlea sound processing system and method with customization function | |
CN110428843A (en) | A kind of voice gender identification deep learning method | |
CN111951824A (en) | Detection method for distinguishing depression based on sound | |
WO2020249532A1 (en) | A neural network model for cochlear mechanics and processing | |
Hüwel et al. | Hearing aid research data set for acoustic environment recognition | |
CN113850013B (en) | Ship radiation noise classification method | |
CN112466284B (en) | Mask voice identification method | |
Peng et al. | An acoustic signal processing system for identification of queen-less beehives | |
CN107221338A (en) | Sound wave extraction element and extracting method | |
CN116092512A (en) | Small sample voice separation method based on data generation | |
CN115862639A (en) | Artificial intelligence voice analysis method based on K-means clustering analysis | |
Kothapally et al. | Speech Detection and Enhancement Using Single Microphone for Distant Speech Applications in Reverberant Environments. | |
CN114938487A (en) | Hearing aid self-fitting method based on sound scene discrimination | |
CN113887339A (en) | Silent voice recognition system and method fusing surface electromyogram signal and lip image | |
CN111144482B (en) | Scene matching method and device for digital hearing aid and computer equipment | |
CN113488071A (en) | Pig cough recognition method, device, equipment and readable storage medium | |
CN113870901B (en) | SVM-KNN-based voice emotion recognition method | |
CN114155878B (en) | Artificial intelligence detection system, method and computer program |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |