Create app.py

app.py

@@ -0,0 +1,48 @@
+from transformers import VitsModel, AutoTokenizer
+import torch
+import gradio as gr
+import spaces
+import numpy as np
+
+device = "cuda" if torch.cuda.is_available() else "cpu"
+print(f"Using device: {device}")
+
+@spaces.GPU
+model = VitsModel.from_pretrained("facebook/mms-tts-eng").to(device)
+tokenizer = AutoTokenizer.from_pretrained("facebook/mms-tts-eng")
+
+@spaces.GPU
+def process_text(text):
+    inputs = tokenizer(text, return_tensors="pt").to(device)
+    
+    with torch.no_grad():
+        output = model(**inputs).waveform
+    
+    audio_numpy = output.cpu().numpy().squeeze()
+    sample_rate = model.config.sampling_rate
+    return (sample_rate, audio_numpy)
+
+examples = [
+    ["Hello, welcome to text-to-speech system!"],
+    ["How amazing is artificial intelligence technology?"],
+    ["The weather is beautiful today, isn't it?"],
+    ["Learning new things makes life exciting."],
+    ["This audio was generated by artificial intelligence."]
+]
+
+with gr.Blocks() as demo:
+    gr.Markdown("## 🎤 MMS-TTS English Text-to-Speech System")
+    gr.Markdown("Enter text below and convert it to natural sounding speech!")
+    
+    with gr.Row():
+        with gr.Column():
+            input_text = gr.Textbox(label="Input Text", placeholder="Enter text here...")
+            gr.Examples(examples=examples, inputs=input_text, label="Example Texts")
+            submit_btn = gr.Button("Generate Speech")
+        
+        with gr.Column():
+            audio_output = gr.Audio(label="Generated Speech", type="numpy")
+    
+    submit_btn.click(fn=process_text, inputs=input_text, outputs=audio_output)
+
+demo.launch()