Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
language: da
|
| 3 |
+
tags:
|
| 4 |
+
- speech
|
| 5 |
+
- xls_r
|
| 6 |
+
- xls_r_pretrained
|
| 7 |
+
- danish
|
| 8 |
+
license: apache-2.0
|
| 9 |
+
---
|
| 10 |
+
## XLS-R-300m-danish
|
| 11 |
+
|
| 12 |
+
Continued pretraining of (facebook/wav2vec2-xls-r-300m)[https://huggingface.co/facebook/wav2vec2-xls-r-300m] for 120.000 steps on 141.000 hours of speech from Danish radio (DR P1 and Radio24Syv from 2005 to 2021).
|
| 13 |
+
|
| 14 |
+
The model was pretrained on 16kHz audio using fairseq and should be fine-tuned to perform speech recognition.
|
| 15 |
+
|
| 16 |
+
The model was trained by (Lasse Hansen)[https://github.com/HLasse] from (CHCAA)[https://chcaa.io] in collaboration with (Alvenir)[https://alvenir.ai] on the (UCloud)[https:/cloud.sdu.dk] platform.
|