53 lines
1.5 KiB
Markdown
53 lines
1.5 KiB
Markdown
---
|
||
language:
|
||
- da
|
||
license: other
|
||
datasets:
|
||
- ftspeech
|
||
metrics:
|
||
- wer
|
||
tasks:
|
||
- automatic-speech-recognition
|
||
base_model: facebook/wav2vec2-xls-r-300m
|
||
model-index:
|
||
- name: wav2vec2-xls-r-300m-ftspeech
|
||
results:
|
||
- task:
|
||
type: automatic-speech-recognition
|
||
dataset:
|
||
name: Danish Common Voice 8.0
|
||
type: mozilla-foundation/common_voice_8_0
|
||
args: da
|
||
metrics:
|
||
- type: wer
|
||
value: 17.91
|
||
- task:
|
||
type: automatic-speech-recognition
|
||
dataset:
|
||
name: Alvenir ASR test dataset
|
||
type: Alvenir/alvenir_asr_da_eval
|
||
metrics:
|
||
- type: wer
|
||
value: 13.84
|
||
---
|
||
|
||
# XLS-R-300m-FTSpeech
|
||
|
||
## Model description
|
||
|
||
This model is a fine-tuned version of [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on the [FTSpeech dataset](https://ftspeech.github.io/), being a dataset of 1,800 hours of transcribed speeches from the Danish parliament.
|
||
|
||
|
||
## Performance
|
||
|
||
The model achieves the following WER scores (lower is better):
|
||
|
||
| **Dataset** | **WER without LM** | **WER with 5-gram LM** |
|
||
| :---: | ---: | ---: |
|
||
| [Danish part of Common Voice 8.0](https://huggingface.co/datasets/mozilla-foundation/common_voice_8_0/viewer/da/train) | 20.48 | 17.91 |
|
||
| [Alvenir test set](https://huggingface.co/datasets/Alvenir/alvenir_asr_da_eval) | 15.46 | 13.84 |
|
||
|
||
|
||
## License
|
||
|
||
The use of this model needs to adhere to [this license from the Danish Parliament](https://www.ft.dk/da/aktuelt/tv-fra-folketinget/deling-og-rettigheder). |