Update README.md

trained on french only material from https://librivox.org/ public domain material and voice.

I have trained on 2 rounds :
1- 80k samples, 100 epochs, 1 single speaker
2-120k samples, 150 epochs, different corpus speakers

You can infer online using https://huggingface.co/spaces/RASPIAUDIO/f5-tts_french

README.md

@@ -1,3 +1,7 @@
----
-license: cc-by-nc-4.0
----
+---
+license: cc-by-nc-4.0
+language:
+- fr
+base_model:
+- SWivid/F5-TTS
+---