Create app.py

app.py

@@ -0,0 +1,69 @@
+import torch
+import gradio as gr
+import speech_recognition as sr
+import pyttsx3
+import time
+from transformers import AutoModelForSeq2SeqLM, AutoTokenizer
+from happytransformer import HappyTextToText, TTSettings  # Using HappyTransformer
+
+def load_models():
+    model_name = "prithivida/grammar_error_correcter_v1"
+    tokenizer = AutoTokenizer.from_pretrained(model_name)
+    model = AutoModelForSeq2SeqLM.from_pretrained(model_name)
+    happy_tt = HappyTextToText("T5", "prithivida/grammar_error_correcter_v1")  # Using T5-based model
+    return tokenizer, model, happy_tt
+
+def transcribe_audio(audio):
+    recognizer = sr.Recognizer()
+    with sr.AudioFile(audio) as source:
+        audio_data = recognizer.record(source)
+    try:
+        text = recognizer.recognize_google(audio_data)
+        return text
+    except sr.UnknownValueError:
+        return "Could not understand the audio."
+    except sr.RequestError as e:
+        return f"Speech recognition error: {e}"
+
+def correct_grammar(text, tokenizer, model, happy_tt):
+    inputs = tokenizer.encode("gec: " + text, return_tensors="pt", max_length=128, truncation=True)
+    with torch.no_grad():
+        outputs = model.generate(inputs, max_length=128, num_return_sequences=1)
+    corrected_text = tokenizer.decode(outputs[0], skip_special_tokens=True)
+    
+    args = TTSettings(num_beams=5, min_length=1)
+    correction = happy_tt.generate_text("gec: " + text, args=args).text  # Better correction method
+    grammar_score = 100 - abs(len(text) - len(correction))  # Scoring based on text change ratio
+    
+    return corrected_text, grammar_score, correction
+
+def gradio_interface(text):
+    tokenizer, model, happy_tt = load_models()
+    corrected_text, grammar_score, correction = correct_grammar(text, tokenizer, model, happy_tt)
+    return corrected_text, grammar_score, correction
+
+def gradio_audio_interface(audio):
+    text = transcribe_audio(audio)
+    return gradio_interface(text)
+
+def main():
+    iface = gr.Interface(
+        fn=gradio_interface,
+        inputs=gr.Textbox(placeholder="Enter a sentence..."),
+        outputs=["text", "number", "text"],
+        title="AI Grammar Checker",
+        description="Enter text to check grammar, get suggestions, and see a score."
+    )
+    
+    audio_iface = gr.Interface(
+        fn=gradio_audio_interface,
+        inputs=gr.Audio(source="microphone", type="filepath"),
+        outputs=["text", "text", "number", "text"],
+        title="AI Grammar Checker (Audio)",
+        description="Speak to check grammar, get suggestions, and see a score."
+    )
+    
+    gr.TabbedInterface([iface, audio_iface], ["Text Input", "Speech Input"]).launch()
+
+if __name__ == "__main__":
+    main()