File size: 3,233 Bytes
70e5c60 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 |
# 📁 Estructura del proyecto generado para 'voice'
# 📦 voice/
# ├── README.md
# ├── requirements.txt
# ├── src/
# │ ├── model.py
# │ ├── train.py
# │ └── inference.py
# ├── data/
# │ └── samples/
# └── app_gradio.py
# ========================================
# ✅ README.md (inicial)
readme = '''
# 🐽 Kuchiuya Voice Synthesizer
Este proyecto permite entrenar e inferir voces sintéticas al estilo Kuchiuya.
## Requisitos
```bash
pip install -r requirements.txt
```
## Entrenamiento
```bash
python src/train.py
```
## Inferencia
```bash
python src/inference.py --text "El barro no miente."
```
## Demo Web
```bash
python app_gradio.py
```
'''
# ========================================
# ✅ requirements.txt
requirements = '''
torch
torchaudio
numpy
scipy
librosa
soundfile
gradio
pyttsx3
'''
# ========================================
# ✅ model.py
model_py = '''
import torch.nn as nn
class SimpleTTSModel(nn.Module):
def __init__(self, input_dim=256, hidden_dim=512, output_dim=80):
super(SimpleTTSModel, self).__init__()
self.model = nn.Sequential(
nn.Linear(input_dim, hidden_dim),
nn.ReLU(),
nn.Linear(hidden_dim, output_dim)
)
def forward(self, x):
return self.model(x)
'''
# ========================================
# ✅ train.py
train_py = '''
import torch
import torch.nn as nn
import torch.optim as optim
from src.model import SimpleTTSModel
import numpy as np
print("🐷 Entrenando modelo Kuchiuya...")
# Datos simulados (para demostrar el flujo)
x = torch.randn((100, 256))
y = torch.randn((100, 80))
model = SimpleTTSModel()
criterion = nn.MSELoss()
optimizer = optim.Adam(model.parameters(), lr=0.001)
for epoch in range(10):
optimizer.zero_grad()
outputs = model(x)
loss = criterion(outputs, y)
loss.backward()
optimizer.step()
print(f"Epoch {epoch+1}/10, Pérdida: {loss.item():.4f}")
# Guardar modelo
torch.save(model.state_dict(), "tts_model.pth")
print("✅ Modelo guardado como 'tts_model.pth'")
'''
# ========================================
# ✅ inference.py
inference_py = '''
import argparse
import pyttsx3
parser = argparse.ArgumentParser()
parser.add_argument('--text', type=str, required=True, help='Texto a sintetizar')
args = parser.parse_args()
engine = pyttsx3.init()
engine.setProperty('rate', 150)
engine.save_to_file(args.text, 'output.wav')
engine.runAndWait()
print(f"🔊 Texto sintetizado: '{args.text}' → guardado como 'output.wav'")
'''
# ========================================
# ✅ app_gradio.py
app_gradio_py = '''
import gradio as gr
import pyttsx3
import os
def sintetizar(texto):
engine = pyttsx3.init()
engine.setProperty('rate', 150)
output_path = "voz_kuchiuya.wav"
engine.save_to_file(texto, output_path)
engine.runAndWait()
return output_path
demo = gr.Interface(
fn=sintetizar,
inputs=gr.Textbox(lines=2, placeholder="Escribe tu frase Kuchiuya aquí..."),
outputs=gr.Audio(type="filepath"),
title="Voz Kuchiuya",
description="Convierte texto en una voz ritual cyberpunk del universo Ariplanteater."
)
demo.launch()
'''
|