Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,45 @@
|
|
| 1 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2 |
license: apache-2.0
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
+
language: pt
|
| 3 |
+
datasets:
|
| 4 |
+
- CORAA
|
| 5 |
+
- common_voice
|
| 6 |
+
- mls
|
| 7 |
+
- cetuc
|
| 8 |
+
- voxforge
|
| 9 |
+
metrics:
|
| 10 |
+
- wer
|
| 11 |
+
tags:
|
| 12 |
+
- audio
|
| 13 |
+
- speech
|
| 14 |
+
- wav2vec2
|
| 15 |
+
- pt
|
| 16 |
+
- portuguese-speech-corpus
|
| 17 |
+
- automatic-speech-recognition
|
| 18 |
+
- speech
|
| 19 |
+
- PyTorch
|
| 20 |
license: apache-2.0
|
| 21 |
+
model-index:
|
| 22 |
+
- name: Alef Iury XLSR Wav2Vec2 Large 53 Portuguese
|
| 23 |
+
results:
|
| 24 |
+
- task:
|
| 25 |
+
name: Speech Recognition
|
| 26 |
+
type: automatic-speech-recognition
|
| 27 |
+
metrics:
|
| 28 |
+
- name: Test CORAA WER
|
| 29 |
+
type: wer
|
| 30 |
+
value: 24.89%
|
| 31 |
---
|
| 32 |
+
|
| 33 |
+
# Wav2vec 2.0 trained with CORAA Portuguese Dataset and Open Portuguese Datasets
|
| 34 |
+
|
| 35 |
+
This a the demonstration of a fine-tuned Wav2vec model for Portuguese using the following datasets:
|
| 36 |
+
|
| 37 |
+
- [CORAA dataset](https://github.com/nilc-nlp/CORAA)
|
| 38 |
+
- [CETUC](http://www02.smt.ufrj.br/~igor.quintanilha/alcaim.tar.gz).
|
| 39 |
+
- [Multilingual Librispeech (MLS)](http://www.openslr.org/94/).
|
| 40 |
+
- [VoxForge](http://www.voxforge.org/).
|
| 41 |
+
- [Common Voice 6.1](https://commonvoice.mozilla.org/pt).
|
| 42 |
+
|
| 43 |
+
## Repository
|
| 44 |
+
|
| 45 |
+
The repository that implements the model to be trained and tested is avaible [here](https://github.com/alefiury/SE-R_2022_Challenge_Wav2vec2).
|