scholarly journals Speech Enhancement for Hearing Aids with Deep Learning on Environmental Noises

2020 ◽  
Vol 10 (17) ◽  
pp. 6077
Author(s):  
Gyuseok Park ◽  
Woohyeong Cho ◽  
Kyu-Sung Kim ◽  
Sangmin Lee
Keyword(s):  
Neural Networks ◽  
Noise Reduction ◽  
Speech Enhancement ◽  
Hearing Aids ◽  
Deep Neural Networks ◽  
Hearing Aid ◽  
Environmental Noise ◽  
Speech Quality ◽  
Impaired Hearing ◽  
Noise Classification

Hearing aids are small electronic devices designed to improve hearing for persons with impaired hearing, using sophisticated audio signal processing algorithms and technologies. In general, the speech enhancement algorithms in hearing aids remove the environmental noise and enhance speech while still giving consideration to hearing characteristics and the environmental surroundings. In this study, a speech enhancement algorithm was proposed to improve speech quality in a hearing aid environment by applying noise reduction algorithms with deep neural network learning based on noise classification. In order to evaluate the speech enhancement in an actual hearing aid environment, ten types of noise were self-recorded and classified using convolutional neural networks. In addition, noise reduction for speech enhancement in the hearing aid were applied by deep neural networks based on the noise classification. As a result, the speech quality based on the speech enhancements removed using the deep neural networks—and associated environmental noise classification—exhibited a significant improvement over that of the conventional hearing aid algorithm. The improved speech quality was also evaluated by objective measure through the perceptual evaluation of speech quality score, the short-time objective intelligibility score, the overall quality composite measure, and the log likelihood ratio score.

scholarly journals Environmental Noise Classification Using Convolutional Neural Networks with Input Transform for Hearing Aids

2020 ◽  
Vol 17 (7) ◽  
pp. 2270 ◽  
Author(s):  
Gyuseok Park ◽  
Sangmin Lee
Keyword(s):  
Neural Networks ◽  
Convolutional Neural Networks ◽  
Hearing Aids ◽  
Classification Accuracy ◽  
Median Filter ◽  
Environmental Noise ◽  
Classification Algorithm ◽  
Correct Classification ◽  
Time Length ◽  
Noise Classification

Hearing aids are essential for people with hearing loss, and noise estimation and classification are some of the most important technologies used in devices. This paper presents an environmental noise classification algorithm for hearing aids that uses convolutional neural networks (CNNs) and image signals transformed from sound signals. The algorithm was developed using the data of ten types of noise acquired from living environments where such noises occur. Spectrogram images transformed from sound data are used as the input of the CNNs after processing of the images by a sharpening mask and median filter. The classification results of the proposed algorithm were compared with those of other noise classification methods. A maximum correct classification accuracy of 99.25% was achieved by the proposed algorithm for a spectrogram time length of 1 s, with the correct classification accuracy decreasing with increasing spectrogram time length up to 8 s. For a spectrogram time length of 8 s and using the sharpening mask and median filter, the classification accuracy was 98.73%, which is comparable with the 98.79% achieved by the conventional method for a time length of 1 s. The proposed hearing aid noise classification algorithm thus offers less computational complexity without compromising on performance.

Performance Improvement of Digital Hearing Aid Systems

2015 ◽  
Vol 1 ◽  
pp. 27
Author(s):  
Isiaka Ajewale Alimi
Keyword(s):  
Noise Reduction ◽  
Hearing Aids ◽  
Digital Signal Processor ◽  
Speech Intelligibility ◽  
Communication Systems ◽  
Hearing Aid ◽  
Spectral Subtraction ◽  
Residual Spectrum ◽  
Digital Hearing Aids ◽  
Digital Hearing Aid

Digital hearing aids addresses the issues of noise and speech intelligibility that is associated with the analogue types. One of the main functions of the digital signal processor (DSP) of digital hearing aid systems is noise reduction which can be achieved by speech enhancement algorithms which in turn improve system performance and flexibility. However, studies have shown that the quality of experience (QoE) with some of the current hearing aids is not up to expectation in a noisy environment due to interfering sound, background noise and reverberation. It is also suggested that noise reduction features of the DSP can be further improved accordingly. Recently, we proposed an adaptive spectral subtraction algorithm to enhance the performance of communication systems and address the issue of associated musical noise generated by the conventional spectral subtraction algorithm. The effectiveness of the algorithm has been confirmed by different objective and subjective evaluations. In this study, an adaptive spectral subtraction algorithm is implemented using the noise-estimation algorithm for highly non-stationary noisy environments instead of the voice activity detection (VAD) employed in our previous work due to its effectiveness. Also, signal to residual spectrum ratio (SR) is implemented in order to control the amplification distortion for speech intelligibility improvement. The results show that the proposed scheme gives comparatively better performance and can be easily employed in digital hearing aid system for improving speech quality and intelligibility.

Text-informed speech enhancement with deep neural networks

2015 ◽  
Author(s):  
Keisuke Kinoshita ◽  
Marc Delcroix ◽  
Atsunori Ogawa ◽  
Tomohiro Nakatani
Keyword(s):  
Neural Networks ◽  
Speech Enhancement ◽  
Deep Neural Networks

scholarly journals Smartphone-Based System for Learning and Inferring Hearing Aid Settings

2016 ◽  
Vol 27 (09) ◽  
pp. 732-749 ◽  
Author(s):  
Gabriel Aldaz ◽  
Sunil Puria ◽  
Larry J. Leifer
Keyword(s):  
Noise Reduction ◽  
Hearing Aids ◽  
User Study ◽  
Hearing Aid ◽  
Sensor Data ◽  
Significant Preference ◽  
Environmental Sound ◽  
Sound Environment ◽  
Study Participants ◽  
Hearing System

Background: Previous research has shown that hearing aid wearers can successfully self-train their instruments’ gain-frequency response and compression parameters in everyday situations. Combining hearing aids with a smartphone introduces additional computing power, memory, and a graphical user interface that may enable greater setting personalization. To explore the benefits of self-training with a smartphone-based hearing system, a parameter space was chosen with four possible combinations of microphone mode (omnidirectional and directional) and noise reduction state (active and off). The baseline for comparison was the “untrained system,” that is, the manufacturer’s algorithm for automatically selecting microphone mode and noise reduction state based on acoustic environment. The “trained system” first learned each individual’s preferences, self-entered via a smartphone in real-world situations, to build a trained model. The system then predicted the optimal setting (among available choices) using an inference engine, which considered the trained model and current context (e.g., sound environment, location, and time). Purpose: To develop a smartphone-based prototype hearing system that can be trained to learn preferred user settings. Determine whether user study participants showed a preference for trained over untrained system settings. Research Design: An experimental within-participants study. Participants used a prototype hearing system—comprising two hearing aids, Android smartphone, and body-worn gateway device—for ˜6 weeks. Study Sample: Sixteen adults with mild-to-moderate sensorineural hearing loss (HL) (ten males, six females; mean age = 55.5 yr). Fifteen had ≥6 mo of experience wearing hearing aids, and 14 had previous experience using smartphones. Intervention: Participants were fitted and instructed to perform daily comparisons of settings (“listening evaluations”) through a smartphone-based software application called Hearing Aid Learning and Inference Controller (HALIC). In the four-week-long training phase, HALIC recorded individual listening preferences along with sensor data from the smartphone—including environmental sound classification, sound level, and location—to build trained models. In the subsequent two-week-long validation phase, participants performed blinded listening evaluations comparing settings predicted by the trained system (“trained settings”) to those suggested by the hearing aids’ untrained system (“untrained settings”). Data Collection and Analysis: We analyzed data collected on the smartphone and hearing aids during the study. We also obtained audiometric and demographic information. Results: Overall, the 15 participants with valid data significantly preferred trained settings to untrained settings (paired-samples t test). Seven participants had a significant preference for trained settings, while one had a significant preference for untrained settings (binomial test). The remaining seven participants had nonsignificant preferences. Pooling data across participants, the proportion of times that each setting was chosen in a given environmental sound class was on average very similar. However, breaking down the data by participant revealed strong and idiosyncratic individual preferences. Fourteen participants reported positive feelings of clarity, competence, and mastery when training via HALIC. Conclusions: The obtained data, as well as subjective participant feedback, indicate that smartphones could become viable tools to train hearing aids. Individuals who are tech savvy and have milder HL seem well suited to take advantages of the benefits offered by training with a smartphone.

Evaluation of a Transient Noise Reduction Strategy for Hearing Aids

2012 ◽  
Vol 23 (08) ◽  
pp. 606-615 ◽  
Author(s):  
HaiHong Liu ◽  
Hua Zhang ◽  
Ruth A. Bentler ◽  
Demin Han ◽  
Luo Zhang
Keyword(s):  
Noise Reduction ◽  
Hearing Aids ◽  
Background Noise ◽  
Hearing Aid ◽  
Sound Quality ◽  
T Test ◽  
Reduction Strategy ◽  
Speech Stimuli ◽  
Significant Difference ◽  
Speech Clarity

Background: Transient noise can be disruptive for people wearing hearing aids. Ideally, the transient noise should be detected and controlled by the signal processor without disrupting speech and other intended input signals. A technology for detecting and controlling transient noises in hearing aids was evaluated in this study. Purpose: The purpose of this study was to evaluate the effectiveness of a transient noise reduction strategy on various transient noises and to determine whether the strategy has a negative impact on sound quality of intended speech inputs. Research Design: This was a quasi-experimental study. The study involved 24 hearing aid users. Each participant was asked to rate the parameters of speech clarity, transient noise loudness, and overall impression for speech stimuli under the algorithm-on and algorithm-off conditions. During the evaluation, three types of stimuli were used: transient noises, speech, and background noises. The transient noises included “knife on a ceramic board,” “mug on a tabletop,” “office door slamming,” “car door slamming,” and “pen tapping on countertop.” The speech sentences used for the test were presented by a male speaker in Mandarin. The background noises included “party noise” and “traffic noise.” All of these sounds were combined into five listening situations: (1) speech only, (2) transient noise only, (3) speech and transient noise, (4) background noise and transient noise, and (5) speech and background noise and transient noise. Results: There was no significant difference on the ratings of speech clarity between the algorithm-on and algorithm-off (t-test, p = 0.103). Further analysis revealed that speech clarity was significant better at 70 dB SLP than 55 dB SPL (p < 0.001). For transient noise loudness: under the algorithm-off condition, the percentages of subjects rating the transient noise to be somewhat soft, appropriate, somewhat loud, and too loud were 0.2, 47.1, 29.6, and 23.1%, respectively. The corresponding percentages under the algorithm-on were 3.0, 72.6, 22.9, and 1.4%, respectively. A significant difference on the ratings of the transient noise loudness was found between the algorithm-on and algorithm-off (t-test, p < 0.001). For overall impression for speech stimuli: under the algorithm-off condition, the percentage of subjects rating the algorithm to be not helpful at all, somewhat helpful, helpful, and very helpful for speech stimuli were 36.5, 20.8, 33.9, and 8.9%, respectively. Under the algorithm-on condition, the corresponding percentages were 35.0, 19.3, 30.7, and 15.0%, respectively. Statistical analysis revealed there was a significant difference on the ratings of overall impression on speech stimuli. The ratings under the algorithm-on condition were significantly more helpful for speech understanding than the ratings under algorithm-off (t-test, p < 0.001). Conclusions: The transient noise reduction strategy appropriately controlled the loudness for most of the transient noises and did not affect the sound quality, which could be beneficial to hearing aid wearers.

Sign in / Sign up

Export Citation Format

Share Document