From 35f130ee85ff7b86283f5ba1e7d80fef6722c598 Mon Sep 17 00:00:00 2001
From: SWivid <swivid@qq.com>
Date: Wed, 4 Jun 2025 06:11:49 +0800
Subject: [PATCH] minor update for infer-gradio

---
 src/f5_tts/infer/infer_gradio.py | 26 +++++++++++++++++---------
 1 file changed, 17 insertions(+), 9 deletions(-)

diff --git a/src/f5_tts/infer/infer_gradio.py b/src/f5_tts/infer/infer_gradio.py
index c6d1d9b..f4c3aef 100644
--- a/src/f5_tts/infer/infer_gradio.py
+++ b/src/f5_tts/infer/infer_gradio.py
@@ -128,7 +128,7 @@ def load_text_from_file(file):
     return gr.update(value=text)
 
 
-@lru_cache(maxsize=100)  # NOTE. need to ensure params of infer() hashable
+@lru_cache(maxsize=1000)  # NOTE. need to ensure params of infer() hashable
 @gpu_decorator
 def infer(
     ref_audio_orig,
@@ -209,14 +209,6 @@ def infer(
     return (final_sample_rate, final_wave), spectrogram_path, ref_text, used_seed
 
 
-with gr.Blocks() as app_credits:
-    gr.Markdown("""
-# Credits
-
-* [mrfakename](https://github.com/fakerybakery) for the original [online demo](https://huggingface.co/spaces/mrfakename/E2-F5-TTS)
-* [RootingInLoad](https://github.com/RootingInLoad) for initial chunk generation and podcast app exploration
-* [jpgallegoar](https://github.com/jpgallegoar) for multiple speech-type generation & voice chat
-""")
 with gr.Blocks() as app_tts:
     gr.Markdown("# Batched TTS")
     ref_audio_input = gr.Audio(label="Reference Audio", type="filepath")
@@ -320,6 +312,12 @@ with gr.Blocks() as app_tts:
         outputs=[ref_text_input],
     )
 
+    ref_audio_input.clear(
+        lambda: [None, None],
+        None,
+        [ref_text_input, ref_text_file],
+    )
+
     generate_btn.click(
         basic_tts,
         inputs=[
@@ -932,6 +930,16 @@ Have a conversation with an AI using your reference voice!
             )
 
 
+with gr.Blocks() as app_credits:
+    gr.Markdown("""
+# Credits
+
+* [mrfakename](https://github.com/fakerybakery) for the original [online demo](https://huggingface.co/spaces/mrfakename/E2-F5-TTS)
+* [RootingInLoad](https://github.com/RootingInLoad) for initial chunk generation and podcast app exploration
+* [jpgallegoar](https://github.com/jpgallegoar) for multiple speech-type generation & voice chat
+""")
+
+
 with gr.Blocks() as app:
     gr.Markdown(
         f"""