Update README.md
Browse files
README.md
CHANGED
|
@@ -15,5 +15,18 @@ Trained over 247+h hours of "train" split of facebook/multilingual_librispeech d
|
|
| 15 |
- catastrophic failure (the model forgot english)
|
| 16 |
- italian pronunciation not perfect
|
| 17 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 18 |
|
| 19 |
The run.py file is an example of how to extract the wav files and produce the metadata.csv to use for training
|
|
|
|
| 15 |
- catastrophic failure (the model forgot english)
|
| 16 |
- italian pronunciation not perfect
|
| 17 |
|
| 18 |
+
## folder structure:
|
| 19 |
+
| - italian_59kh
|
| 20 |
+
| | - checkpoints
|
| 21 |
+
|
| 22 |
+
### italian_59kh
|
| 23 |
+
Contains the weight at specific steps, the higher the number, the further it went into training.
|
| 24 |
+
Weights in this folder can't be used to resume training, use checkpoints instead.
|
| 25 |
+
|
| 26 |
+
### italian_59kh/checkpoints
|
| 27 |
+
Contains the weight of the checkpoints at specific steps, the higher the number, the further it went into training.
|
| 28 |
+
Weights in this folder can be used as starting point to continue training.
|
| 29 |
+
|
| 30 |
+
|
| 31 |
|
| 32 |
The run.py file is an example of how to extract the wav files and produce the metadata.csv to use for training
|