Organisation Name

Leibniz University Hannover

Industry

Education

Organisation Website

https://www.uni-hannover.de/de/

Country

Germany

Sustainable Development Goals (SDGs)

General Description of the AI tool

The aim of Phonomatics is to develop an automatic speech recognition (ASR) software that transcribes child speech recordings from natural communication situations (e.g., looking at a picture book with another child or adult) and evaluates them linguistically (vocabulary, grammar, articulation). The software output will be various speech and language measures aggregated for different professional users and assessment purpuses. Therefore, our software can be used in educational and health care settings (primarly at kindergarten age) wherever child language development is being monitored.

Github, open data repository

https://www.tnt.uni-hannover.de/en/project/talc/

Relevant Research and Publications

1. Rumberg, L., Gebauer, C., Ehlert, H., Wallbaum, M., Bornholt, L., Ostermann, J. & Lüdtke, U. (2022). kidsTALC: A Corpus of 3- to 11-year-old German Children’s Connected Natural Speech. Proceedings INTERSPEECH — 23th Annual Conference of the International Speech Communication Association, 5160-5164. (Our open access database to promote research on child language technology)

2. Christopher Gebauer, Lars Rumberg, Hanna Ehlert, Ulrike Lüdtke, Jörn Ostermann
Exploiting Diversity of Automatic Transcripts from Distinct Speech Recognition Techniques for Children’s Speech
Proceedings INTERSPEECH 2023 – 24th Annual Conference of the International Speech Communication Association, August 2023
(pdf) BibTeX (Work on ASR of child speech as a prerequisite for a detailed speech assessment)

3. Rumberg, L., Gebauer, C., Ehlert, H., Wallbaum, M., Lüdtke, U. & Ostermann, J. (2023). Uncertainty Estimation for Connectionist Temporal Classification Based Automatic Speech Recognition. Proceedings INTERSPEECH, 4583-4587. (Uncertainty estimation as a method to provide transparent automated outputs)

4. Christopher Gebauer, Lars Rumberg, Lars Köhn, Hanna Ehlert, Edith Beaulac, Jörn Ostermann
Grammatical Error Detection on Spontaneous Children’s Speech Using Iterative Pseudo Labeling
to appear in Proceedings INTERSPEECH 2025 – 26th Annual Conference of the International Speech Communication Association, ISCA, 2025 (exploration of automated linguistic assessment)

5. Lüdtke, U., Bornman, J., de Wet, F., Heid, U., Ostermann, J., Rumberg, L., van der Linde, J., & Ehlert, H. (2023). Multi-disciplinary perspectives on automatic analysis of children's language samples: Where do we go from here? Folia Phoniatirca et Logopaedica, 75, 1-12. (Interdiscuplinary, international review on child speech technology with the joint German and South African team)

Needs