Papers by Carlos-D. Martínez-Hinarejos
Comparison of Conventional Hybrid and CTC/Attention Decoders for Continuous Visual Speech Recognition (2024.lrec-main)
Copied to clipboard
| Challenge: | Recent advances have been achieved in Visual Speech Recognition (VSR) despite the lack of data, there is no clear comparison between different types of decoders for certain languages and tasks. |
| Approach: | They focused on how the conventional DNN-HMM decoder behaves depending on the amount of data used for their estimation. |
| Outcome: | The proposed model improves the CTC/Attention model in data-scarcity scenarios while requiring less training time and fewer parameters. |
AnnoTheia: A Semi-Automatic Annotation Toolkit for Audio-Visual Speech Technologies (2024.lrec-main)
Copied to clipboard
| Challenge: | a small fraction of the languages currently covered by speech technologies are mainly spoken in English. |
| Approach: | They present an annotation toolkit that detects when a person speaks on the scene and the corresponding transcription. |
| Outcome: | The proposed toolkit can speed up the annotation process by up to four times . it can be used in Spanish, and is available on github. |
LIP-RTVE: An Audiovisual Database for Continuous Spanish in the Wild (2022.lrec-1)
Copied to clipboard
| Challenge: | Speech perception is considered as a purely auditory process, but it is a multi-modal process involving multiple senses. |
| Approach: | They propose to use a semi-automatically annotated audiovisual database to deal with unconstrained natural Spanish. |
| Outcome: | The proposed system can be used to estimate speech recognition systems in the Deep Learning era. |