from flask import Flask, request, jsonify
from transformers import pipeline
from transformers import AutoTokenizer, AutoModelForTokenClassification
import whisper
import os
import tempfile
import io
import torchaudio

app = Flask(__name__)

# Initialize Whisper model
whisper_model = whisper.load_model("small")  # Renamed variable

# Initialize Emotion Classifier
classifier = pipeline("text-classification", model="j-hartmann/emotion-english-distilroberta-base", return_all_scores=True)

# Initialize NER pipeline
ner_tokenizer = AutoTokenizer.from_pretrained("dslim/bert-base-NER")
ner_model = AutoModelForTokenClassification.from_pretrained("dslim/bert-base-NER")  # Renamed variable
ner_pipeline = pipeline("ner", model=ner_model, tokenizer=ner_tokenizer)  # Renamed variable


@app.route('/transcribe', methods=['POST'])
def transcribe():
    try:
        # Read raw bytes from the request
        audio_bytes = request.data  
        if not audio_bytes:
            return jsonify({"error": "No audio data provided"}), 400

        # Convert bytes to a file-like object
        audio_file = io.BytesIO(audio_bytes)

        # Load audio as a waveform using torchaudio
        waveform, sample_rate = torchaudio.load(audio_file)

        # Whisper expects a NumPy array, so we convert it
        audio_numpy = waveform.squeeze().numpy()

        # Transcribe the audio
        result = model.transcribe(audio_numpy)

        return jsonify({"text": result["text"]})

    except Exception as e:
        print("Error:", str(e))  # Log error for debugging
        return jsonify({"error": "Internal Server Error", "details": str(e)}), 500

        
        
@app.route('/classify', methods=['POST'])
def classify():
    try:
        data = request.get_json()
        if 'text' not in data:
            return jsonify({"error": "Missing 'text' field"}), 400
        
        text = data['text']
        result = classifier(text)
        return jsonify(result)
    
    except Exception as e:
        return jsonify({"error": str(e)}), 500

@app.route('/ner', methods=['POST'])
def ner_endpoint():
    try:
        data = request.get_json()
        text = data.get("text", "")
        
        # Use the renamed ner_pipeline
        ner_results = ner_pipeline(text)
        
        words_and_entities = [
            {"word": result['word'], "entity": result['entity']} 
            for result in ner_results
        ]
        
        return jsonify({"entities": words_and_entities})
    except Exception as e:
        return jsonify({"error": str(e)}), 500