asr-africa/wolof-1-hour-wav2vec2-xls-r-google-fleurs
Browse files- README.md +31 -2
- model.safetensors +1 -1
README.md
CHANGED
|
@@ -4,9 +4,26 @@ license: apache-2.0
|
|
| 4 |
base_model: facebook/wav2vec2-xls-r-300m
|
| 5 |
tags:
|
| 6 |
- generated_from_trainer
|
|
|
|
|
|
|
|
|
|
|
|
|
| 7 |
model-index:
|
| 8 |
- name: wolof-1-hour-wav2vec2-xls-r-google-fleurs
|
| 9 |
-
results:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 10 |
---
|
| 11 |
|
| 12 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
@@ -15,7 +32,11 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/asr-africa-research-team/ASR%20Africa/runs/ogk3j7c0)
|
| 16 |
# wolof-1-hour-wav2vec2-xls-r-google-fleurs
|
| 17 |
|
| 18 |
-
This model is a fine-tuned version of [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on
|
|
|
|
|
|
|
|
|
|
|
|
|
| 19 |
|
| 20 |
## Model description
|
| 21 |
|
|
@@ -45,6 +66,14 @@ The following hyperparameters were used during training:
|
|
| 45 |
- lr_scheduler_warmup_ratio: 0.1
|
| 46 |
- num_epochs: 50
|
| 47 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 48 |
### Framework versions
|
| 49 |
|
| 50 |
- Transformers 4.44.2
|
|
|
|
| 4 |
base_model: facebook/wav2vec2-xls-r-300m
|
| 5 |
tags:
|
| 6 |
- generated_from_trainer
|
| 7 |
+
datasets:
|
| 8 |
+
- fleurs
|
| 9 |
+
metrics:
|
| 10 |
+
- wer
|
| 11 |
model-index:
|
| 12 |
- name: wolof-1-hour-wav2vec2-xls-r-google-fleurs
|
| 13 |
+
results:
|
| 14 |
+
- task:
|
| 15 |
+
name: Automatic Speech Recognition
|
| 16 |
+
type: automatic-speech-recognition
|
| 17 |
+
dataset:
|
| 18 |
+
name: fleurs
|
| 19 |
+
type: fleurs
|
| 20 |
+
config: wo_sn
|
| 21 |
+
split: None
|
| 22 |
+
args: wo_sn
|
| 23 |
+
metrics:
|
| 24 |
+
- name: Wer
|
| 25 |
+
type: wer
|
| 26 |
+
value: 1.0
|
| 27 |
---
|
| 28 |
|
| 29 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 32 |
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/asr-africa-research-team/ASR%20Africa/runs/ogk3j7c0)
|
| 33 |
# wolof-1-hour-wav2vec2-xls-r-google-fleurs
|
| 34 |
|
| 35 |
+
This model is a fine-tuned version of [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on the fleurs dataset.
|
| 36 |
+
It achieves the following results on the evaluation set:
|
| 37 |
+
- Loss: 3.0203
|
| 38 |
+
- Wer: 1.0
|
| 39 |
+
- Cer: 1.0
|
| 40 |
|
| 41 |
## Model description
|
| 42 |
|
|
|
|
| 66 |
- lr_scheduler_warmup_ratio: 0.1
|
| 67 |
- num_epochs: 50
|
| 68 |
|
| 69 |
+
### Training results
|
| 70 |
+
|
| 71 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
|
| 72 |
+
|:-------------:|:-----:|:----:|:---------------:|:---:|:---:|
|
| 73 |
+
| 4.3753 | 25.0 | 200 | 3.0125 | 1.0 | 1.0 |
|
| 74 |
+
| 3.003 | 50.0 | 400 | 3.0203 | 1.0 | 1.0 |
|
| 75 |
+
|
| 76 |
+
|
| 77 |
### Framework versions
|
| 78 |
|
| 79 |
- Transformers 4.44.2
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1261959180
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f04774d1ca991be9311bd1db407f6afb6c876edcc2ff3d76974810ea2140adf7
|
| 3 |
size 1261959180
|