Matches in SemOpenAlex for { <https://semopenalex.org/work/W4324356870> ?p ?o ?g. }
Showing items 1 to 61 of
61
with 100 items per page.
- W4324356870 abstract "Recorded speech signals often contain noise that affects the quality of the signal and reduces intelligibility. Several studies have used Generative Adversarial Networks (GANs) to remove noise artifacts and improve speech intelligibility. However, GANs can suffer from gradient vanishing or gradient explosion that can reduce their effectiveness in denoising. To mitigate gradient vanishing, we applied the CoulombGAN architecture to speech denoising using a model structure similar to Hifi-GAN, the current state of the art speech denoiser. We call this new model Hifi-CoGAN. We used a WaveNet generator to denoise signals, a PostNet for general cleanup, and a Multi-Resolution Discriminator to evaluate the signal quality relative to the clean signal. Our results show that Hifi-CoGAN was able to outperform Hifi-GAN in many of the narrowband signals (signals with a limited range of frequencies) in terms of the Short-Term Objective Intelligibility (STOI) and Perceptual Evaluation of Speech Quality (PESQ) metrics. However, the model did not perform as well as Hifi-GAN with wideband noise signals (signals with a wider range of frequencies) such as white noise, so future work must be done to improve the model for these noise signals." @default.
- W4324356870 created "2023-03-16" @default.
- W4324356870 creator A5023567582 @default.
- W4324356870 creator A5079077661 @default.
- W4324356870 date "2022-08-31" @default.
- W4324356870 modified "2023-10-01" @default.
- W4324356870 title "Denoising Speech Signals with Hifi-Coulomb-GANs" @default.
- W4324356870 doi "https://doi.org/10.47611/jsrhs.v11i3.3501" @default.
- W4324356870 hasPublicationYear "2022" @default.
- W4324356870 type Work @default.
- W4324356870 citedByCount "0" @default.
- W4324356870 crossrefType "journal-article" @default.
- W4324356870 hasAuthorship W4324356870A5023567582 @default.
- W4324356870 hasAuthorship W4324356870A5079077661 @default.
- W4324356870 hasBestOaLocation W43243568701 @default.
- W4324356870 hasConcept C103734657 @default.
- W4324356870 hasConcept C111472728 @default.
- W4324356870 hasConcept C121332964 @default.
- W4324356870 hasConcept C138885662 @default.
- W4324356870 hasConcept C154945302 @default.
- W4324356870 hasConcept C163294075 @default.
- W4324356870 hasConcept C24890656 @default.
- W4324356870 hasConcept C2776182073 @default.
- W4324356870 hasConcept C2779803651 @default.
- W4324356870 hasConcept C28490314 @default.
- W4324356870 hasConcept C41008148 @default.
- W4324356870 hasConcept C60048801 @default.
- W4324356870 hasConcept C76155785 @default.
- W4324356870 hasConcept C94915269 @default.
- W4324356870 hasConceptScore W4324356870C103734657 @default.
- W4324356870 hasConceptScore W4324356870C111472728 @default.
- W4324356870 hasConceptScore W4324356870C121332964 @default.
- W4324356870 hasConceptScore W4324356870C138885662 @default.
- W4324356870 hasConceptScore W4324356870C154945302 @default.
- W4324356870 hasConceptScore W4324356870C163294075 @default.
- W4324356870 hasConceptScore W4324356870C24890656 @default.
- W4324356870 hasConceptScore W4324356870C2776182073 @default.
- W4324356870 hasConceptScore W4324356870C2779803651 @default.
- W4324356870 hasConceptScore W4324356870C28490314 @default.
- W4324356870 hasConceptScore W4324356870C41008148 @default.
- W4324356870 hasConceptScore W4324356870C60048801 @default.
- W4324356870 hasConceptScore W4324356870C76155785 @default.
- W4324356870 hasConceptScore W4324356870C94915269 @default.
- W4324356870 hasIssue "3" @default.
- W4324356870 hasLocation W43243568701 @default.
- W4324356870 hasOpenAccess W4324356870 @default.
- W4324356870 hasPrimaryLocation W43243568701 @default.
- W4324356870 hasRelatedWork W2037635165 @default.
- W4324356870 hasRelatedWork W2140344044 @default.
- W4324356870 hasRelatedWork W2140410589 @default.
- W4324356870 hasRelatedWork W2542098180 @default.
- W4324356870 hasRelatedWork W3038255278 @default.
- W4324356870 hasRelatedWork W3092562885 @default.
- W4324356870 hasRelatedWork W3111611403 @default.
- W4324356870 hasRelatedWork W3203814202 @default.
- W4324356870 hasRelatedWork W4200562864 @default.
- W4324356870 hasRelatedWork W4375869276 @default.
- W4324356870 hasVolume "11" @default.
- W4324356870 isParatext "false" @default.
- W4324356870 isRetracted "false" @default.
- W4324356870 workType "article" @default.