This is a Piper TTS model finetuned from
Kristin mediumThis model is after just 5 epochs on ~30% of total data I curated (synthetic + natural).
Currently, I'm refining the synthetic dataset as I'm not satisfied with its quality. I will resume finetuning after.
Also running ablations on the best ratio of synthetic and natural data.
From initial observations it seems like its better to use majority of one kind (90%-10%).
Trying to push the boundaries of audio generated by a mere 63 MB model.
1import wave
2
3from src.python_run.piper import PiperVoice # Or import from the installed package if you used pip
4
5model = PiperVoice.load("en_US-ceylia-medium.onnx")
6
7text = "I have a big plan for today. It involves fine-tuning you."
8
9with wave.open("output.wav", "wb") as output_file:
10 output_file.setnchannels(1)
11 output_file.setsampwidth(2)
12 output_file.setframerate(22050)
13 model.synthesize(text=text, wav_file=output_file, sentence_silence=0.25)
Bryce Beattie for training the Kristin model.
Reference Audio from datasets by
@Jinsaryko
