Natural TTS Synthesis by Conditioning WaveNet on Mel Spectrogram Predictions
Paper • 1712.05884 • Published • 3
Japanese, male single-speaker Tacotron2 TTS model trained with double decoder consistency (DDC) on the Kokoro dataset.
Trained by @erogol and originally published at: https://github.com/coqui-ai/TTS/releases/v0.6.1_models/
With Coqui TTS:
from TTS.api import TTS
tts = TTS("tts_models/ja/kokoro/tacotron2-DDC")
tts.tts_to_file("こんにちは", file_path="output.wav")