German ASR Transformer model trained on CoVoST2 using espnet2
Abstract
German ASR transformer model trained on CoVoST2 using espnet2 Results are as below: ### WER |dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err||---|---|---|---|---|---|---|---|---| |dev.de-en|13505|120056|85.1|13.2|1.8|2.0|16.9|65.7||test.de-en|13502|118600|83.0|14.9|2.1|2.0|19.1|68.4| ### CER |dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err||---|---|---|---|---|---|---|---|---| |dev.de-en|13505|819319|96.2|2.1|1.7|1.1|5.0|65.7||test.de-en|13502|801401|95.3|2.6|2.1|1.4|6.1|68.4| ### TER |dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err||---|---|---|---|---|---|---|---|---| |dev.de-en|13505|289208|88.2|8.0|3.8|1.1|12.9|65.7||test.de-en|13502|282287|86.6|9.3|4.1|1.4|14.9|68.4|
Fields of Study & Subjects
Speech Recognition and Synthesis
Natural Language Processing Techniques
Speech and Audio Processing
Publication details
- Journal / source
- Zenodo (CERN European Organization for Nuclear Research)
- Publication date
- 12 January 2026
- Document type
- Dataset