import os
import pandas as pd
import numpy as np
import librosa
import soundfile as sf
from tqdm import tqdm
from concurrent.futures import ProcessPoolExecutor

class AudioConverter:
    def __init__(self, csv_path, audio_dir, output_dir, sampling_rate=16000):
        self.csv_path = csv_path
        self.audio_dir = audio_dir
        self.output_dir = output_dir
        self.sampling_rate = sampling_rate

    def process_row(self, row):
        source, noise, snr, caption = row['source'], row['noise'], int(row['snr']), row['caption']

        # Load source and noise audio files
        source_path = os.path.join(self.audio_dir, f'{source}.wav')
        noise_path = os.path.join(self.audio_dir, f'{noise}.wav')

        source_audio, _ = librosa.load(source_path, sr=self.sampling_rate, mono=True)
        noise_audio, _ = librosa.load(noise_path, sr=self.sampling_rate, mono=True)

        # Create audio mixture with a specific SNR level
        source_power = np.mean(source_audio ** 2)
        noise_power = np.mean(noise_audio ** 2)
        desired_noise_power = source_power / (10 ** (snr / 10))
        scaling_factor = np.sqrt(desired_noise_power / noise_power)
        noise_audio *= scaling_factor

        mixture = source_audio + noise_audio

        # Declip if necessary
        max_value = np.max(np.abs(mixture))
        if max_value > 1:
            source_audio *= 0.9 / max_value
            mixture *= 0.9 / max_value

        # Save the mixture to a new file
        output_path = os.path.join(self.output_dir, f'{source}_{noise}_{snr}.wav')
        sf.write(output_path, mixture, self.sampling_rate)

        return f'Saved mixture to {output_path}'

    def convert(self):
        # Read the CSV file
        data = pd.read_csv(self.csv_path)

        # Ensure the output directory exists
        os.makedirs(self.output_dir, exist_ok=True)

        with ProcessPoolExecutor(max_workers=8) as executor:
            results = list(tqdm(executor.map(self.process_row, [row for _, row in data.iterrows()]), total=len(data)))

        for result in results:
            print(result)

if __name__ == "__main__":
    csv_path = '/gpfs/work4/0/einf6190/audio-datasets/data/LASS/lass_synthetic_validation.csv'
    audio_dir = '/scratch-shared/gwijngaard/data/LASS/lass_validation'
    output_dir = '/gpfs/work4/0/einf6190/audio-datasets/data/LASS/synthetic_validation'

    converter = AudioConverter(csv_path, audio_dir, output_dir)
    converter.convert()