Create app.py

app.py

@@ -0,0 +1,44 @@
+import edge_tts
+import gradio as gr
+import tempfile
+
+language_dict = {
+'English-Jenny (Female)': 'en-US-JennyNeural',
+}
+
+async def text_to_speech_edge(text, language_code, rate, volume, pitch):
+    voice = language_dict.get(language_code, "default_voice")
+    
+    rates = "+" + str(rate) + "%" if rate >= 0 else str(rate) + "%"
+    volumes = "+" + str(volume) + "%" if volume >= 0 else str(volume) + "%"
+    pitchs = "+" + str(pitch) + "Hz" if pitch >= 0 else str(pitch) + "Hz"
+
+    communicate = edge_tts.Communicate(text, voice, rate=rates, volume=volumes, pitch=pitchs, proxy=None)
+
+    with tempfile.NamedTemporaryFile(delete=False, suffix=".wav") as tmp_file:
+        tmp_path = tmp_file.name
+        await communicate.save(tmp_path)
+    return f"Speech synthesis completed for: {text}", tmp_path
+
+input_text = gr.Textbox(lines=5, label="Input Text")
+output_text = gr.Textbox(label="Output Text")
+output_audio = gr.Audio(type="filepath", label="Exported Audio")
+language = gr.Dropdown(choices=list(language_dict.keys()), label="Choose the Voice Model")
+rate = gr.Slider(-100, 100,step=1,value=0,label="Rate",info="Rate",interactive=True)
+
+volume = gr.Slider(-100,100,step=1,value=0,label="Volume",info="Volume",interactive=True)
+
+pitch = gr.Slider(-100,100,step=1,value=0,label="Pitch",info="Pitch",interactive=True)
+
+
+
+interface = gr.Interface(
+    fn=text_to_speech_edge,
+    inputs=[input_text, language, rate, volume,pitch],
+    outputs=[output_text, output_audio],
+    title="Edge-TTS",
+    description="Microsoft Edge Text-To-Speech (Created by Yash Chouhan)",
+
+)
+
+interface.launch(share=True)