From a7fd2e7e9a612f3fb94744b2cdfacbf5efa65eb7 Mon Sep 17 00:00:00 2001
From: lpscr <147736764+lpscr@users.noreply.github.com>
Date: Mon, 28 Oct 2024 07:09:37 +0200
Subject: [PATCH] finetune quick note for ema (#298)

* add note about ema
---
 src/f5_tts/train/finetune_gradio.py | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/src/f5_tts/train/finetune_gradio.py b/src/f5_tts/train/finetune_gradio.py
index 7136253..1b42c6a 100644
--- a/src/f5_tts/train/finetune_gradio.py
+++ b/src/f5_tts/train/finetune_gradio.py
@@ -1472,6 +1472,9 @@ If you encounter a memory error, try reducing the batch size per GPU to a smalle
             )
 
         with gr.TabItem("test model"):
+            gr.Markdown("""```plaintext 
+SOS : check the use_ema setting (True or False) for your model to see what works best for you. 
+```""")
             exp_name = gr.Radio(label="Model", choices=["F5-TTS", "E2-TTS"], value="F5-TTS")
             list_checkpoints, checkpoint_select = get_checkpoints_project(projects_selelect, False)