main.py

# main.py

from fastapi import FastAPI, File, UploadFile, HTTPException
from fastapi.responses import JSONResponse
from pydantic import BaseModel
import librosa
import numpy as np
import tempfile
import os
import warnings
import matplotlib.pyplot as plt

warnings.filterwarnings("ignore", category=UserWarning, module='librosa')

app = FastAPI()

def extract_audio_features(audio_file_path):
    # Load the audio file and extract features
    y, sr = librosa.load(audio_file_path, sr=None)
    f0, voiced_flag, voiced_probs = librosa.pyin(y, fmin=75, fmax=600)
    f0 = f0[~np.isnan(f0)]
    energy = librosa.feature.rms(y=y)[0]
    mfccs = librosa.feature.mfcc(y=y, sr=sr, n_mfcc=13)
    onset_env = librosa.onset.onset_strength(y=y, sr=sr)
    tempo, _ = librosa.beat.beat_track(onset_envelope=onset_env, sr=sr)
    speech_rate = tempo / 60
    return f0, energy, speech_rate, mfccs, y, sr

def analyze_voice_stress(audio_file_path):
    f0, energy, speech_rate, mfccs, y, sr = extract_audio_features(audio_file_path)
    mean_f0 = np.mean(f0)
    std_f0 = np.std(f0)
    mean_energy = np.mean(energy)
    std_energy = np.std(energy)
    gender = 'male' if mean_f0 < 165 else 'female'
    norm_mean_f0 = 110 if gender == 'male' else 220
    norm_std_f0 = 20
    norm_mean_energy = 0.02
    norm_std_energy = 0.005
    norm_speech_rate = 4.4
    norm_std_speech_rate = 0.5
    z_f0 = (mean_f0 - norm_mean_f0) / norm_std_f0
    z_energy = (mean_energy - norm_mean_energy) / norm_std_energy
    z_speech_rate = (speech_rate - norm_speech_rate) / norm_std_speech_rate
    stress_score = (0.4 * z_f0) + (0.4 * z_speech_rate) + (0.2 * z_energy)
    stress_level = float(1 / (1 + np.exp(-stress_score)) * 100)
    categories = ["Very Low Stress", "Low Stress", "Moderate Stress", "High Stress", "Very High Stress"]
    category_idx = min(int(stress_level / 20), 4)
    stress_category = categories[category_idx]
    return {"stress_level": stress_level, "category": stress_category, "gender": gender}

class StressResponse(BaseModel):
    stress_level: float
    category: str
    gender: str

@app.post("/analyze-stress/", response_model=StressResponse)
async def analyze_stress(file: UploadFile = File(...)):
    if not file.filename.endswith(".wav"):
        raise HTTPException(status_code=400, detail="Only .wav files are supported.")
    with tempfile.NamedTemporaryFile(delete=False, suffix=".wav") as temp_file:
        temp_file.write(await file.read())
        temp_file_path = temp_file.name
    try:
        result = analyze_voice_stress(temp_file_path)
        return JSONResponse(content=result)
    except Exception as e:
        raise HTTPException(status_code=500, detail=str(e))
    finally:
        os.remove(temp_file_path)

if __name__ == "__main__":
    import uvicorn
    uvicorn.run("main:app", host="0.0.0.0", port=7860, reload=True)