51 lines
2.3 KiB
Python
51 lines
2.3 KiB
Python
import os
|
|
import sys
|
|
import warnings
|
|
# Suppress warnings from tensorflow and other libraries
|
|
warnings.filterwarnings("ignore", category=UserWarning)
|
|
warnings.filterwarnings("ignore", category=FutureWarning)
|
|
def main():
|
|
import argparse
|
|
parser = argparse.ArgumentParser(description="IndexTTS Command Line")
|
|
parser.add_argument("text", type=str, help="Text to be synthesized")
|
|
parser.add_argument("-v", "--voice", type=str, required=True, help="Path to the audio prompt file (wav format)")
|
|
parser.add_argument("-o", "--output_path", type=str, default="gen.wav", help="Path to the output wav file")
|
|
parser.add_argument("-c", "--config", type=str, default="checkpoints/config.yaml", help="Path to the config file. Default is 'checkpoints/config.yaml'")
|
|
parser.add_argument("--model_dir", type=str, default="checkpoints", help="Path to the model directory. Default is 'checkpoints'")
|
|
parser.add_argument("--fp16", action="store_true", default=True, help="Use FP16 for inference if available")
|
|
parser.add_argument("-f", "--force", action="store_true", default=False, help="Force to overwrite the output file if it exists")
|
|
args = parser.parse_args()
|
|
|
|
if not os.path.exists(args.voice):
|
|
print(f"Audio prompt file {args.voice} does not exist.")
|
|
parser.print_help()
|
|
sys.exit(1)
|
|
if not os.path.exists(args.config):
|
|
print(f"Config file {args.config} does not exist.")
|
|
parser.print_help()
|
|
sys.exit(1)
|
|
|
|
output_path = args.output_path
|
|
if os.path.exists(output_path):
|
|
if not args.force:
|
|
print(f"ERROR: Output file {output_path} already exists. Use --force to overwrite.")
|
|
parser.print_help()
|
|
sys.exit(1)
|
|
else:
|
|
os.remove(output_path)
|
|
|
|
try:
|
|
import torch
|
|
if not torch.cuda.is_available():
|
|
print("WARNING: CUDA is not available. Running on CPU may be slow.")
|
|
except ImportError:
|
|
print("ERROR: PyTorch is not installed. Please install it first.")
|
|
sys.exit(1)
|
|
|
|
|
|
from indextts.infer import IndexTTS
|
|
tts = IndexTTS(cfg_path=args.config, model_dir=args.model_dir, is_fp16=args.fp16)
|
|
tts.infer(audio_prompt=args.voice, text=args.text, output_path=output_path)
|
|
|
|
if __name__ == "__main__":
|
|
main() |