File size: 2,414 Bytes
10ea2f8 393129e 10ea2f8 393129e 10ea2f8 |
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 |
"""
Chiluka TTS - PyTorch Hub Example
Load the model using torch.hub.load() - no pip install needed,
just PyTorch and a GitHub repo.
Requirements:
pip install torch torchaudio
sudo apt-get install espeak-ng
Usage:
python torchhub_example.py --reference path/to/reference.wav
python torchhub_example.py --reference ref.wav --variant telugu --language te
"""
import argparse
import torch
def main():
parser = argparse.ArgumentParser(description="Chiluka TTS - PyTorch Hub Example")
parser.add_argument("--reference", type=str, required=True, help="Path to reference audio file")
parser.add_argument("--variant", type=str, default="default", choices=["default", "telugu", "hindi_english"],
help="Model variant (default, telugu, hindi_english)")
parser.add_argument("--text", type=str, default=None, help="Text to synthesize")
parser.add_argument("--language", type=str, default=None, help="Language code (en-us, hi, te)")
parser.add_argument("--output", type=str, default="output_torchhub.wav", help="Output wav file path")
args = parser.parse_args()
# Set defaults
if args.text is None:
if args.variant == "telugu":
args.text = "నమస్కారం, నేను చిలుక మాట్లాడుతున్నాను"
else:
args.text = "Hello, I am Chiluka, a text to speech system."
if args.language is None:
if args.variant == "telugu":
args.language = "te"
else:
args.language = "en-us"
# Load via torch.hub
# Available entry points:
# 'chiluka' - Hindi-English model (default)
# 'chiluka_telugu' - Telugu model
# 'chiluka_hindi_english' - Hindi-English model (explicit)
print(f"Loading model via torch.hub (variant: {args.variant})...")
if args.variant == "telugu":
tts = torch.hub.load('Seemanth/chiluka', 'chiluka_telugu')
else:
tts = torch.hub.load('Seemanth/chiluka', 'chiluka')
# Synthesize
print(f"Synthesizing: '{args.text}'")
print(f"Language: {args.language}")
wav = tts.synthesize(
text=args.text,
reference_audio=args.reference,
language=args.language,
)
# Save
tts.save_wav(wav, args.output)
print(f"Duration: {len(wav) / 24000:.2f} seconds")
if __name__ == "__main__":
main()
|