Delete train.py

train.py

@@ -1,54 +0,0 @@
-import pandas as pd
-import tensorflow as tf
-import numpy as np
-import librosa
-import os
-
-DATA_PATH = "data/transcriptions.csv"
-AUDIO_DIR = "data"
-MODEL_PATH = "model/clone_tts_model.h5"
-SAMPLE_RATE = 22050
-TEXT_MAX_LEN = 100  # Max characters per text
-
-# === Load and preprocess dataset ===
-def load_data():
-    data = pd.read_csv(DATA_PATH)
-    texts = data['text'].values
-    audio_arrays = []
-
-    for file in data['file']:
-        audio_path = os.path.join(AUDIO_DIR, file)
-        y, _ = librosa.load(audio_path, sr=SAMPLE_RATE)
-        audio_arrays.append(y)
-
-    max_audio_len = max(len(a) for a in audio_arrays)
-    padded_audios = np.array([np.pad(a, (0, max_audio_len - len(a))) for a in audio_arrays])
-
-    padded_texts = np.array([
-        [ord(c) for c in text.ljust(TEXT_MAX_LEN)[:TEXT_MAX_LEN]] for text in texts
-    ])
-
-    return padded_texts, padded_audios, max_audio_len
-
-# === Build and train model ===
-def train_model():
-    print("Loading and preparing data...")
-    X, y, audio_len = load_data()
-
-    print("Building model...")
-    model = tf.keras.Sequential([
-        tf.keras.layers.Input(shape=(TEXT_MAX_LEN,)),
-        tf.keras.layers.Dense(256, activation='relu'),
-        tf.keras.layers.Dense(audio_len)
-    ])
-
-    model.compile(optimizer='adam', loss='mse')
-    print("Training...")
-    model.fit(X, y, epochs=10, batch_size=4)
-
-    os.makedirs(os.path.dirname(MODEL_PATH), exist_ok=True)
-    model.save(MODEL_PATH)
-    print(f"Model saved to {MODEL_PATH}")
-
-if __name__ == "__main__":
-    train_model()