German to English MT Transformer model trained on CoVoST2 using Espnet2
Abstract
# RESULTS## Environments- date: `Tue Jan 13 08:31:44 CET 2026`- python version: `3.8.20 | packaged by conda-forge | (default, Sep 30 2024, 17:52:49) [GCC 13.3.0]`- espnet version: `espnet 202503`- pytorch version: `pytorch 2.0.0` ## mt_train_mt_raw_bpe_lc.rm1000 ### BLEU |dataset|score|verbose_score||---|---|---||decode_mt_mt_model_valid.acc.ave/dev.de-en|30.6|63.6/39.4/25.6/17.3 (BP = 0.942 ratio = 0.944 hyp_len = 126403 ref_len = 133922)||decode_mt_mt_model_valid.acc.ave/test.de-en|30.3|63.4/39.1/25.4/17.0 (BP = 0.943 ratio = 0.944 hyp_len = 124317 ref_len = 131673)|
Fields of Study & Subjects
Natural Language Processing Techniques
Speech Recognition and Synthesis
Topic Modeling
Publication details
- Journal / source
- Zenodo (CERN European Organization for Nuclear Research)
- Publication date
- 12 January 2026
- Document type
- Dataset