Using recurrent neural networks to improve the perception of speech in non-stationary noise by people with cochlear implants.

@article{Goehring2019UsingRN,
  title={Using recurrent neural networks to improve the perception of speech in non-stationary noise by people with cochlear implants.},
  author={Tobias Goehring and Mahmoud Keshavarzi and Robert P. Carlyon and Brian C. J. Moore},
  journal={The Journal of the Acoustical Society of America},
  year={2019},
  volume={146 1},
  pages={
          705
        }
}
Speech-in-noise perception is a major problem for users of cochlear implants (CIs), especially with non-stationary background noise. Noise-reduction algorithms have produced benefits but relied on a priori information about the target speaker and/or background noise. A recurrent neural network (RNN) algorithm was developed for enhancing speech in non-stationary noise and its benefits were evaluated for speech perception, using both objective measures and experiments with CI simulations and CI… 

Figures and Tables from this paper

Transient Noise Reduction Using a Deep Recurrent Neural Network: Effects on Subjective Speech Intelligibility and Listening Comfort
TLDR
A deep recurrent neural network for reducing transient sounds was developed and its effects on subjective speech intelligibility and listening comfort were investigated, with significantly better results for hearing-impaired participants.
Improving the Intelligibility of Speech for Simulated Electric and Acoustic Stimulation Using Fully Convolutional Neural Networks
  • N. Wang, H. Wang, Yu Tsao
  • Physics
    IEEE Transactions on Neural Systems and Rehabilitation Engineering
  • 2021
TLDR
This study, being the first to evaluate deep learning SE approaches for EAS, confirms that FCN(S) is an effective SE approach that may potentially be integrated into an EAS processor to benefit users in noisy environments.
Vibro-Tactile Enhancement of Speech Intelligibility in Multi-talker Noise for Simulated Cochlear Implant Listening
TLDR
Presentation of tactile stimulation of speech in multi-talker noise could be achieved by a compact, portable device and offer an inexpensive and noninvasive means for improving speech-in-noise performance in CI users.
Deep Learning-Based Speech Enhancement With a Loss Trading Off the Speech Distortion and the Noise Residue for Cochlear Implants
The cochlea plays a key role in the transmission from acoustic vibration to neural stimulation upon which the brain perceives the sound. A cochlear implant (CI) is an auditory prosthesis to replace
Using Spectral Blurring to Assess Effects of Channel Interaction on Speech-in-Noise Perception with Cochlear Implants
TLDR
Spectral blurring is introduced to simulate some of the effects of channel interaction by adjusting the overlap between electrode channels at the input level of the analysis filters or at the output by using several simultaneously stimulated electrodes per channel to improve speech-in-noise perception by CI listeners.
A New Approach for Noise Suppression in Cochlear Implants: A Single-Channel Noise Reduction Algorithm1
TLDR
The monaural signal-to-noise ratio estimation-based noise suppression algorithm “eVoice,” which is incorporated in the processors of Nurotron® EnduroTM, was evaluated in two speech perception experiments and shows that speech intelligibility in stationary speech-shaped noise can be significantly improved with eVoice.
The effect of input noises on the activity of auditory neurons using GLM-based metrics
TLDR
It is found that non-stationary noise clearly contributes to the multi-unit neural activity in the IC by causing excitation, regardless of the SNR, input level or vocalization type, which indicates that the so-called noise invariance in theIC is dependent on the input noisy conditions.
Comparison of Speech Recognition in Cochlear Implant Users with Different Speech Processors.
TLDR
CI users had a lower signal-to-noise ratio and a higher percentage of sentence recognition with the OTE processor than with the BTE processor, and both sound processors performed well in all noise conditions.
The effect of increased channel interaction on speech perception with cochlear implants
TLDR
Results confirm and extend earlier findings indicating that CI speech perception may not benefit from deactivating individual channels along the array and that efforts should instead be directed towards reducing channel interaction per se and in particular for the most-apical electrodes.
An Optimal Envelope-Based Noise Reduction Method for Cochlear Implants: An Upper Bound Performance Investigation
TLDR
A new method for noise reduction, especially designed for CI, based on the minimization of the mean square error between the squared envelopes of the estimated and target speech is proposed, which may result in an intelligibility increase of up to 70%, for SNR = −25 dB, as compared to the WF.
...
1
2
3
4
...

References

SHOWING 1-10 OF 63 REFERENCES
Speech enhancement based on neural networks applied to cochlear implant coding strategies
TLDR
A speech enhancement algorithm integrating an artificial neural network (NN) into CI coding strategies is proposed, which decomposes the noisy input signal into time-frequency units, extracts a set of auditory-inspired features and feeds them to the NN to produce an estimation of which CI channels contain more perceptually important information.
Use of a Deep Recurrent Neural Network to Reduce Wind Noise: Effects on Judged Speech Intelligibility and Sound Quality
TLDR
Reduction of wind noise using an RNN is possible and might have beneficial effects when used in hearing aids, and there was a preference for RNN over HPF.
Large-scale training to increase speech intelligibility for hearing-impaired listeners in novel noises.
TLDR
The results indicate that DNN-based supervised speech segregation with large-scale training is a very promising approach for generalization to new acoustic environments.
An algorithm to improve speech recognition in noise for hearing-impaired listeners.
TLDR
Testing using normal-hearing and HI listeners indicated that intelligibility increased following processing in all conditions, and increases were larger for HI listeners, for the modulated background, and for the least-favorable SNRs.
Comparison of effects on subjective intelligibility and quality of speech in babble for two algorithms: A deep recurrent neural network and spectral subtraction.
TLDR
Objective computational measures of speech intelligibility predicted better intelligibility for RNN than for SS or NP, and Processing using the RNN was significantly preferred over NP and over SS processing for both subjective intelligibility and sound quality, although the magnitude of the preferences was small.
An algorithm to increase speech intelligibility for hearing-impaired listeners in novel segments of the same noise type.
TLDR
Substantial sentence-intelligibility benefit was observed for hearing-impaired listeners in both noise types, despite the use of unseen noise segments during the test stage, which highlights the importance of evaluating these algorithms not only in human subjects, but in members of the actual target population.
Speech Recognition in Noise for Cochlear Implantees with a Two-Microphone Monaural Adaptive Noise Reduction System
Objective In this study the performance of a noise reduction strategy applied to cochlear implants is evaluated. The noise reduction strategy is based on a 2-channel adaptive filtering strategy using
Environment-specific noise suppression for improved speech intelligibility by cochlear implant users.
  • Y. Hu, P. Loizou
  • Physics
    The Journal of the Acoustical Society of America
  • 2010
TLDR
The present study demonstrated that the environment-specific approach to noise reduction has the potential to restore speech intelligibility in noise to a level near to that attained in quiet.
An algorithm that improves speech intelligibility in noise for normal-hearing listeners.
TLDR
The findings from this study suggest that algorithms that can estimate reliably the SNR in each T-F unit can improve speech intelligibility.
...
1
2
3
4
5
...