annotation

app.py

@@ -3,7 +3,7 @@ import os
 import soundfile as sf
 import numpy as np
 
-# 这是你现有的推理管线
+
 from decode import InferencePipeline
 from datahandler import AudioMixer, fix_audio_format
 from omegaconf import OmegaConf
@@ -73,7 +73,7 @@ def gradio_TSE(input_audio_path, enroll_audio_path1, enroll_audio_path2, audio_t
 with gr.Blocks() as demo:
     gr.Markdown("## Target Speaker Extraction Demo")
     gr.Markdown(
-        "This demo can handle either clean audio (which we'll turn into a mix) or directly a mix audio."
+        "This demo can handle either clean audio (which we'll turn into a mix) or directly a mix audio. Due to limited training data, recording directly with a mobile browser is more likely to achieve good results"
     )
 
     with gr.Row():