Upload 2 files

app.py

@@ -0,0 +1,35 @@
+import gradio as gr
+import torch
+from transformers import pipeline
+
+# Load models
+speech_to_text = pipeline("automatic-speech-recognition", model="openai/whisper-small")
+translation = pipeline("translation", model="Helsinki-NLP/opus-mt-hi-en")
+text_to_speech = pipeline("text-to-speech", model="espnet/kan-bayashi_ljspeech_vits")
+
+def translate_speech(audio):
+    # Step 1: Hindi speech to Hindi text
+    hindi_text = speech_to_text(audio)["text"]
+    
+    # Step 2: Hindi text to English translation
+    english_text = translation(hindi_text)[0]["translation_text"]
+    
+    # Step 3: English text to speech
+    english_speech = text_to_speech(english_text)["audio"]
+    
+    return hindi_text, english_text, (48000, english_speech)
+
+# Gradio Interface
+demo = gr.Interface(
+    fn=translate_speech,
+    inputs=gr.Audio(source="microphone", type="filepath"),
+    outputs=[
+        gr.Textbox(label="Hindi Text"),
+        gr.Textbox(label="English Translation"),
+        gr.Audio(label="English Audio")
+    ],
+    title="Hindi to English Speech Translator",
+    description="🎤 Speak in Hindi and get the English translation spoken aloud"
+)
+
+demo.launch()

requirements.txt

@@ -0,0 +1,3 @@
+transformers
+torch
+gradio