Matches in SemOpenAlex for { <https://semopenalex.org/work/W4366204855> ?p ?o ?g. }
Showing items 1 to 61 of
61
with 100 items per page.
- W4366204855 abstract "Speech separation remains an important area of multi-speaker signal processing. Deep neural network (DNN) models have attained the best performance on many speech separation benchmarks. Some of these models can take significant time to train and have high memory requirements. Previous work has proposed shortening training examples to address these issues but the impact of this on model performance is not yet well understood. In this work, the impact of applying these training signal length (TSL) limits is analysed for two speech separation models: SepFormer, a transformer model, and Conv-TasNet, a convolutional model. The WJS0-2Mix, WHAMR and Libri2Mix datasets are analysed in terms of signal length distribution and its impact on training efficiency. It is demonstrated that, for specific distributions, applying specific TSL limits results in better performance. This is shown to be mainly due to randomly sampling the start index of the waveforms resulting in more unique examples for training. A SepFormer model trained using a TSL limit of 4.42s and dynamic mixing (DM) is shown to match the best-performing SepFormer model trained with DM and unlimited signal lengths. Furthermore, the 4.42s TSL limit results in a 44% reduction in training time with WHAMR." @default.
- W4366204855 created "2023-04-19" @default.
- W4366204855 creator A5004802529 @default.
- W4366204855 creator A5027797344 @default.
- W4366204855 creator A5030528300 @default.
- W4366204855 date "2023-04-14" @default.
- W4366204855 modified "2023-10-14" @default.
- W4366204855 title "On Data Sampling Strategies for Training Neural Network Speech Separation Models" @default.
- W4366204855 doi "https://doi.org/10.48550/arxiv.2304.07142" @default.
- W4366204855 hasPublicationYear "2023" @default.
- W4366204855 type Work @default.
- W4366204855 citedByCount "0" @default.
- W4366204855 crossrefType "posted-content" @default.
- W4366204855 hasAuthorship W4366204855A5004802529 @default.
- W4366204855 hasAuthorship W4366204855A5027797344 @default.
- W4366204855 hasAuthorship W4366204855A5030528300 @default.
- W4366204855 hasBestOaLocation W43662048551 @default.
- W4366204855 hasConcept C119857082 @default.
- W4366204855 hasConcept C134306372 @default.
- W4366204855 hasConcept C151201525 @default.
- W4366204855 hasConcept C153180895 @default.
- W4366204855 hasConcept C154945302 @default.
- W4366204855 hasConcept C199360897 @default.
- W4366204855 hasConcept C2776864781 @default.
- W4366204855 hasConcept C2779843651 @default.
- W4366204855 hasConcept C28490314 @default.
- W4366204855 hasConcept C33923547 @default.
- W4366204855 hasConcept C41008148 @default.
- W4366204855 hasConcept C50644808 @default.
- W4366204855 hasConcept C51632099 @default.
- W4366204855 hasConcept C81363708 @default.
- W4366204855 hasConceptScore W4366204855C119857082 @default.
- W4366204855 hasConceptScore W4366204855C134306372 @default.
- W4366204855 hasConceptScore W4366204855C151201525 @default.
- W4366204855 hasConceptScore W4366204855C153180895 @default.
- W4366204855 hasConceptScore W4366204855C154945302 @default.
- W4366204855 hasConceptScore W4366204855C199360897 @default.
- W4366204855 hasConceptScore W4366204855C2776864781 @default.
- W4366204855 hasConceptScore W4366204855C2779843651 @default.
- W4366204855 hasConceptScore W4366204855C28490314 @default.
- W4366204855 hasConceptScore W4366204855C33923547 @default.
- W4366204855 hasConceptScore W4366204855C41008148 @default.
- W4366204855 hasConceptScore W4366204855C50644808 @default.
- W4366204855 hasConceptScore W4366204855C51632099 @default.
- W4366204855 hasConceptScore W4366204855C81363708 @default.
- W4366204855 hasLocation W43662048551 @default.
- W4366204855 hasOpenAccess W4366204855 @default.
- W4366204855 hasPrimaryLocation W43662048551 @default.
- W4366204855 hasRelatedWork W2175746458 @default.
- W4366204855 hasRelatedWork W2613736958 @default.
- W4366204855 hasRelatedWork W2732542196 @default.
- W4366204855 hasRelatedWork W2738221750 @default.
- W4366204855 hasRelatedWork W2760085659 @default.
- W4366204855 hasRelatedWork W2883200793 @default.
- W4366204855 hasRelatedWork W3012978760 @default.
- W4366204855 hasRelatedWork W3027997911 @default.
- W4366204855 hasRelatedWork W3093612317 @default.
- W4366204855 hasRelatedWork W4287776258 @default.
- W4366204855 isParatext "false" @default.
- W4366204855 isRetracted "false" @default.
- W4366204855 workType "article" @default.