Update README.md
Browse files
README.md
CHANGED
|
@@ -7,4 +7,13 @@ base_model:
|
|
| 7 |
- SWivid/F5-TTS
|
| 8 |
pipeline_tag: text-to-speech
|
| 9 |
license: cc-by-4.0
|
| 10 |
-
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 7 |
- SWivid/F5-TTS
|
| 8 |
pipeline_tag: text-to-speech
|
| 9 |
license: cc-by-4.0
|
| 10 |
+
---
|
| 11 |
+
|
| 12 |
+
This is a test to see how to finetune F5 in italian
|
| 13 |
+
|
| 14 |
+
Trained over 9h split of facebook/multilingual_librispeech dataset for 200 Epoch:
|
| 15 |
+
- catastrophic failure (the model forgot english)
|
| 16 |
+
- lost ability to clone voice properly
|
| 17 |
+
- italian pronunciation not yet good enough
|
| 18 |
+
|
| 19 |
+
The last produced file, the one to test, is model_italian_200e_9h.safetensors
|