Background noise classification using random forest tree classifier for cochlear implant applications (original) (raw)
Related papers
Real-Time Automatic Tuning of Noise Suppression Algorithms for Cochlear Implant Applications
IEEE Transactions on Biomedical Engineering, 2012
The performance of cochlear implants deteriorates 5 in noisy environments compared to quiet conditions. This paper 6 presents an adaptive cochlear implant system, which is capable of 7 classifying the background noise environment in real time for the 8 purpose of adjusting or tuning its noise suppression algorithm to 9 that environment. The tuning is done automatically with no user 10 intervention. Five objective quality measures are used to show the 11 superiority of this adaptive system compared to a conventional 12 fixed noise-suppression system. Steps taken to achieve the real-13 time implementation of the entire system, incorporating both the 14 cochlear implant speech processing and the background noise sup-15 pression, on a portable PDA research platform are presented along 16 with the timing results. 17 Index Terms-Automatic tuning of noise suppression, charac-18 terization of noisy environments, noise adaptive cochlear implants, 19 real-time implementation of cochlear implant speech processing. 20 I. INTRODUCTION 21 M ORE than 118 000 people around the world have re-22 ceived cochlear implants (CIs) [1]. Since the introduc-23 tion of CIs in 1984, their performance in terms of speech in-24 telligibility has considerably improved. However, their perfor-25 mance in noisy environments still remains a challenge. Speech 26 understanding with cochlear implants is reportedly good in 27 quiet environments but is shown to greatly degrade in noisy 28 environments [2], [3]. Several speech enhancement algorithms, 29 e.g., [4], [5], have been proposed in the literature to address the 30 performance gap in noisy environments. However, no real-time 31 strategy has been offered to automatically tune these algorithms 32 in order to obtain improved performance across different kinds 33 of background noise environments encountered in daily lives by 34 CI patients.
European Archives of Oto-Rhino-Laryngology, 2020
Purpose Refinement currently offered in new sound processors may improve noise listening capability reducing constant background noise and enhancing listening in challenging signal-to-noise conditions. This study aimed to identify whether the new version of speech processor preprocessing strategy contributes to speech recognition in background noise compared to the previous generation processor. Methods This was a multicentric prospective cross-sectional study. Post-lingually deaf adult patients, with at least 1 year of device use and speech recognition scores above 60% on HINT sentences in quiet were invited. Speech recognition performance in quiet and in noise with sound processors with previous and recent technologies was assessed under four conditions with speech coming from the front: (a) quiet (b) fixed noise coming from the front, (c) fixed noise coming from the back, and (d) adaptive noise ratios with noise coming from the front. Results Forty-seven cochlear implant users were included. No significant difference was found in quiet condition. Performance with the new processor was statistically better than the previous sound processor in all three noisy conditions (p < 0.05). With fixed noise coming from the back condition, speech recognition was 62.9% with the previous technology and 73.5% on the new one (p < 0.05). The mean speech recognition in noise was also statistically higher, with 5.8 dB and 7.1 dB for the newer and older technologies (p < 0.05), respectively. Conclusion New technology has shown to provide benefits regarding speech recognition in noise. In addition, the new background noise reduction technology, has shown to be effective and improves speech recognition in situations of more intense noise coming from behind.
Random forest algorithm for improving the performance of speech/non-speech detection
2014 First International Conference on Computational Systems and Communications (ICCSC), 2014
Speech/non-speech detection (SND) distinguishes between speech and non-speech segments in recorded audio and video documents. SND systems can help reduce the storage space required when only speech segments from the audio documents are required, for example content analysis, spoken language identification, etc. In this work, we experimented with the use of time domain, frequency domain and cepstral domain features for short time frames of 20 ms. size along with their mean and standard deviation for segments of size 200 ms. We then analysed if selecting a subset of the features can help improve the performance of the SND system. Towards this, we experimented with different feature selection algorithms, and observed that correlation based feature selection gave the best results. Further, we experimented with different decision tree classification algorithms, and note that random forest algorithm outperformed other decision tree algorithms. We further improved the SND system performance by smoothing the decisions over 5 segments of 200 ms. each. Our baseline system has 272 features, a classification accuracy of 94.45 % and the final system with 8 features has a classification accuracy of 97.80 %.
Cochlear Implant (CI) is the technology, which provides solutions for different types of hearing loss. There is different challenges face by the designers of Cochlear implant in developing signal processing techniques so that can provide the voice which function like normal cochlea of inner ear. This paper discusses various signal processing and neural network techniques used for processing speech data in CI in chronological order of development. The paper also reviews the existing CI devices, their development and techniques used for speech processing in existing CI devices. Wavelet analysis, PET, fMRI are a few techniques used for feature extractions, ambient noise removal and spectral estimation. There are also various other techniques which we will be discussing here. We will also be discussing the development of some environment-specific noise suppression algorithms. Even though there have been some developments in the field of cochlear implant this is an area that is seeing rapid growth. This is the zone of our further discussion.
Investigation on the Effect of the Input Features in the Noise Level Classification of Noisy Speech
2019
Noise Level Estimation plays a crucial role in Speech Enhancement (SE) Algorithms. Recently, few noise estimation (NE) algorithms are developed for SE using the minimal-tracking method, but there is little research done in the noise level classification (NLC). Therefore, there is a need to identify appropriate audio features that are required for the NLC. In this paper, this problem has been addressed and seventeen audio features of the noisy speech are examined for NLC using four different types of standard and efficient classifiers such as K-Nearest Neighbor (KNN), Naive Bayes (NB), Support Vector Machine (SVM), Decision Tree (DT) classifiers. The features are first optimized to achieve the best classification performance using the Principal Component Analysis (PCA) and the Neighbourhood Component Feature Selection (NCFS) method. Finally, a comparative performance analysis is carried out by taking six different categories of real-life noisy speech signals from the standard speech ...
Evaluation of sound classification algorithms for hearing aid applications
2010 IEEE International Conference on Acoustics, Speech and Signal Processing, 2010
Automatic program switching has been shown to be greatly beneficial for hearing aid users. This feature is mediated by a sound classification system, which is traditionally implemented using simple features and heuristic classification schemes, resulting in an unsatisfactory performance in complex auditory scenarios. In this study, a number of experiments are conducted to systematically assess the impact of more sophisticated classifiers and features on automatic acoustic environment classification performance. The results show that advanced classifiers, such as Hidden Markov Model (HMM) or Gaussian Mixture Model (GMM), greatly improve classification performance over simple classifiers. This change does not require a great increase of computational complexity, provided that a suitable number (5 to 7) of low-level features are carefully chosen. These findings indicate that advanced classifiers can be feasible in hearing aid applications.
Investigation on Machine Learning Approaches for Environmental Noise Classifications
Journal of Electrical and Computer Engineering, 2023
Tis project aims to investigate the best machine learning (ML) algorithm for classifying sounds originating from the environment that were considered noise pollution in smart cities. Sound collection was carried out using necessary sound capture tools, after which ML classifcation models were utilized for sound recognition. Additionally, noise pollution monitoring using Python was conducted to provide accurate results for sixteen diferent types of noise that were collected in sixteen cities in Malaysia. Te numbers on the diagonal represent the correctly classifed noises from the test set. Using these correlation matrices, the F1 score was calculated, and a comparison was performed for all models. Te best model was found to be random forest.
Acoustic data classification using random forest algorithm and feed forward neural network
International Journal of Engineering & Technology, 2020
Speaker identification systems are designed to recognize the speaker or set of speakers according to their acoustic analysis. Many approach-es are made to perform the acoustic analysis in the speech signal, the general description of those systems is time and frequency domain analysis. In this paper, acoustic information is extracted from the speech signals using MFCC and Fundamental Frequency methods combi-nation. The results are classified using two different algorithms such as Random-forest and Feed Forward Neural Network. The FFNN classifier integration with the acoustic model resulted a recognition accuracy of 91.4 %. The CMU ARCTIC Database is referred in this work.
Conference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual Conference, 2010
Cochlear implant patients often complain about their difficulty in understanding speech in noisy environments. Currently a fixed noise suppression algorithm is used in cochlear implants regardless of the characteristics of the speech or noise environment. Access to an intelligent mechanism to determine the noise environment on-the-fly in order to automatically switch between different noise suppression algorithms in real-time can enhance patients experience with cochlear implants. In this paper, we report the first prototype system implementing such a real-time switching mechanism for automatic selection between two noise suppression algorithms designed for two commonly encountered noisy environments. The results obtained indicate the feasibility of this on-the-fly switching for actual deployment in cochlear implants.
A COMPARISON OF TECHNIQUES FOR AUDIO ENVIRONMENT CLASSIFICATION
Excessive background noise is one of the most common complaints from hearing aid users. Background noise classification systems can be used in hearing aids to adjust the response based on the noise environment. This paper examines and compares several classification techniques in the form of the k-nearest neighbours (K-NN)classifier, the non-windowed artificial neural network (ANN) and the hidden Markov model (HMM), to an artificial neural network using windowed input (WANN). Results obtained indicate that the WANN gives an accuracy of up to 97.9%, which is the highest accuracy of the tested classifiers. The memory and computational requirements of the windowed ANN are also small compared to the HMM and K-NN. Overall, the WANN is able to give excellent accuracy and reliability and is considered to be a good choice for background noise classification in hearing aids.