Create app.py

app.py

@@ -0,0 +1,21 @@
+import gradio as gr
+from transformers import pipeline
+from transformers import MBartForConditionalGeneration, MBart50TokenizerFast
+transcribe = pipeline("automatic-speech-recognition")
+def speech_to_text(audio):
+    text = transcribe(audio)["text"]
+
+    model_inputs = tokenizer(text, return_tensors="pt", padding=True, truncation=True)
+    generated_tokens = model.generate(
+    **model_inputs,
+    forced_bos_token_id=tokenizer.lang_code_to_id["hi_IN"]
+    )
+
+    translation = tokenizer.batch_decode(generated_tokens, skip_special_tokens=True)
+
+    return translation
+    
+gr.Interface(
+    fn=speech_to_text, 
+    inputs=gr.Audio(source="microphone", type="filepath"), 
+    outputs="text").launch()