Assessing the Generalization Gap of Learning-Based Speech Enhancement Systems in Noisy and Reverberant Environments

Gonzalez, Philippe; Alstrøm, Tommy Sonne; May, Tobias

doi:10.1109/taslp.2023.3318965

Cited by 6 publications

(1 citation statement)

References 73 publications

(97 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A larger and more diverse dataset is key to improving algorithm generalizability. Indeed, the performance of algorithms tends to be overestimated when trained and evaluated on congruent data sets or simplified testing scenarios, such as WSJ-2Mix (mixture of 2 utterances from WSJ0 corpus) 35 . Another important aspect of generalizability is algorithm architecture.…”

Section: Introductionmentioning

confidence: 99%

Deep learning restores speech intelligibility in multi-talker interference for cochlear implant users

Borjigin,

Kokkinakis,

Bharadwaj

et al. 2024

Sci Rep

View full text Add to dashboard Cite

Cochlear implants (CIs) do not offer the same level of effectiveness in noisy environments as in quiet settings. Current single-microphone noise reduction algorithms in hearing aids and CIs only remove predictable, stationary noise, and are ineffective against realistic, non-stationary noise such as multi-talker interference. Recent developments in deep neural network (DNN) algorithms have achieved noteworthy performance in speech enhancement and separation, especially in removing speech noise. However, more work is needed to investigate the potential of DNN algorithms in removing speech noise when tested with listeners fitted with CIs. Here, we implemented two DNN algorithms that are well suited for applications in speech audio processing: (1) recurrent neural network (RNN) and (2) SepFormer. The algorithms were trained with a customized dataset ($$\sim$$ ∼ 30 h), and then tested with thirteen CI listeners. Both RNN and SepFormer algorithms significantly improved CI listener’s speech intelligibility in noise without compromising the perceived quality of speech overall. These algorithms not only increased the intelligibility in stationary non-speech noise, but also introduced a substantial improvement in non-stationary noise, where conventional signal processing strategies fall short with little benefits. These results show the promise of using DNN algorithms as a solution for listening challenges in multi-talker noise interference.

show abstract

Section: Introductionmentioning

confidence: 99%