Papers by Elmar Noeth
Common Phone: A Multilingual Dataset for Robust Acoustic Modelling (2022.lrec-1)
Copied to clipboard
| Challenge: | Current state-of-the-art acoustic models can easily comprise more than 100 million parameters. |
| Approach: | They propose to train a gender-balanced, multilingual corpus from 76.000 contributors via Mozilla’s Common Voice project to perform phonetic symbol recognition and validate the quality of the generated phonetic annotation. |
| Outcome: | The proposed model can perform phonetic symbol recognition and validate the quality of the generated phonetic annotation. |
KSoF: The Kassel State of Fluency Dataset – A Therapy Centered Dataset of Stuttering (2022.lrec-1)
Copied to clipboard
| Challenge: | Stuttering is a complex speech disorder that negatively affects an individual’s ability to communicate effectively. |
| Approach: | They present a therapy-based dataset that tracks stuttering events and changes in speech over a long time and labeled them with six syllable-related event types: blocks, prolongations, sound repetitions, word repetitions and interjections. |
| Outcome: | The study introduces the Kassel State of Fluency (KSoF) dataset containing over 5500 clips of people who underwent stuttering therapy. |