Views
No views yet
| Voice | Language | Quality | Format |
|---|---|---|---|
ta_IN-ValluvaNeural-medium | Tamil (India) | Medium | ONNX |
<model_name> with the name of the voice model you wish to use (e.g., ta_IN-ValluvaNeural-medium).1echo "வணக்கம், எப்படி இருக்கிறீர்கள்?" | piper \
2 --model <model_name>/<model_name>.onnx.onnx \
3 --config <model_name>/<model_name>.onnx.json \
4 --output_file output.wavta_IN-ValluvaNeural-medium:1echo "வணக்கம், எப்படி இருக்கிறீர்கள்?" | piper \
2 --model ta_IN-ValluvaNeural-medium/ta_IN-ValluvaNeural-medium.onnx.onnx \
3 --config ta_IN-ValluvaNeural-medium/ta_IN-ValluvaNeural-medium.onnx.json \
4 --output_file output.wav1import wave
2from piper import PiperVoice
3
4# Replace with your chosen model name
5model_name = "ta_IN-ValluvaNeural-medium"
6model_path = f"{model_name}/{model_name}.onnx.onnx"
7config_path = f"{model_name}/{model_name}.onnx.json"
8
9voice = PiperVoice.load(model_path, config_path=config_path)
10
11text = "வணக்கம், எப்படி இருக்கிறீர்கள்?"
12
13with wave.open("output.wav", "wb") as wav_file:
14 voice.synthesize(text, wav_file)ta_IN-ValluvaNeural-medium was trained up to 2204 epochs and 1388260 steps.<model_name>/
<model_name>.onnx.onnx: The exported ONNX model.<model_name>.onnx.json: The model configuration file.*.ckpt: The training checkpoint file.*.jsonl: The dataset used for training.