Create app.py

app.py

@@ -0,0 +1,28 @@
+# app.py
+
+import gradio as gr
+import torch
+from transformers import pipeline
+
+# Load a fast automatic speech recognition pipeline
+asr_pipeline = pipeline("automatic-speech-recognition", model="facebook/wav2vec2-base-960h")
+
+def transcribe_audio(audio):
+    if audio is None:
+        return "No audio input"
+    
+    text = asr_pipeline(audio)["text"]
+    return text
+
+# Gradio Interface
+iface = gr.Interface(
+    fn=transcribe_audio,
+    inputs=gr.Audio(sources=["microphone"], type="filepath"),
+    outputs=gr.Textbox(label="Recognized Text"),
+    live=True,
+    title="Real-time Voice to Text (Fast Version)",
+    description="Speak into your microphone and get instant transcription!",
+)
+
+if __name__ == "__main__":
+    iface.launch()