Spaces:
Runtime error
Runtime error
File size: 4,461 Bytes
b179e87 |
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 |
import pygame
import uuid
from pathlib import Path
import threading
import edge_tts
import asyncio
from gtts import gTTS
import shutil
import time
class AudioUtils:
AVAILABLE_MODELS = {
'EDGE': {
'name': "es-MX-JorgeNeural", # Voz en español mexicano
'description': "Voz de Edge TTS",
'type': 'cloud',
'fallback': 'gTTS'
},
'VITS': {
'name': "tts_models/es/css10/vits",
'description': "Voz masculina clara y natural",
'type': 'local',
'fallback': 'gTTS'
},
'gTTS': {
'name': "google_tts",
'description': "Google Text-to-Speech",
'type': 'cloud'
}
}
def __init__(self, model_name='EDGE'):
self.is_speaking = False
self.should_stop = False
self.temp_dir = Path("static/temp_audio")
self.temp_dir.mkdir(parents=True, exist_ok=True)
self.current_model = model_name
self.play_lock = threading.Lock()
self.init_audio()
self.cleanup_old_files()
def init_audio(self):
try:
pygame.init()
pygame.mixer.init(frequency=44100, size=-16, channels=2, buffer=4096)
return True
except Exception as e:
print(f"Error inicializando audio: {e}")
return False
async def generate_edge_tts(self, text, output_file):
try:
communicate = edge_tts.Communicate(text, self.AVAILABLE_MODELS['EDGE']['name'])
await communicate.save(str(output_file))
return True
except Exception as e:
print(f"Error con Edge TTS: {e}")
return False
def text_to_speech(self, text, return_file=False):
if not text:
return None
try:
filename = f"{uuid.uuid4()}"
temp_file = self.temp_dir / filename
print(f"Generando audio con modelo {self.current_model}")
if self.current_model == 'EDGE':
temp_file = temp_file.with_suffix('.mp3')
# Ejecutar Edge TTS de manera asíncrona
loop = asyncio.new_event_loop()
asyncio.set_event_loop(loop)
success = loop.run_until_complete(self.generate_edge_tts(text, temp_file))
loop.close()
if not success:
print("Fallback a gTTS")
temp_file = temp_file.with_suffix('.mp3')
tts = gTTS(text=text, lang='es', slow=False)
tts.save(str(temp_file))
else: # gTTS como fallback por defecto
temp_file = temp_file.with_suffix('.mp3')
tts = gTTS(text=text, lang='es', slow=False)
tts.save(str(temp_file))
if not temp_file.exists():
raise Exception(f"El archivo no se generó: {temp_file}")
print(f"Archivo generado exitosamente: {temp_file}")
return temp_file.name if return_file else temp_file
except Exception as e:
print(f"Error crítico en text_to_speech: {e}")
import traceback
traceback.print_exc()
return None
def play_audio(self, file_path):
try:
with self.play_lock:
if pygame.mixer.music.get_busy():
pygame.mixer.music.stop()
pygame.mixer.music.load(str(file_path))
pygame.mixer.music.play()
while pygame.mixer.music.get_busy():
if self.should_stop:
pygame.mixer.music.stop()
self.should_stop = False
break
pygame.time.Clock().tick(10)
except Exception as e:
print(f"Error reproduciendo audio: {e}")
def stop_audio(self):
self.should_stop = True
def cleanup_old_files(self, max_age_minutes=5):
try:
current_time = time.time()
for file in self.temp_dir.glob("*"):
if file.is_file():
file_age_minutes = (current_time - file.stat().st_mtime) / 60
if file_age_minutes > max_age_minutes:
file.unlink()
except Exception as e:
print(f"Error limpiando archivos temporales: {e}") |