Training with custom Dataset

Could you please give me information about training mozillatts with my own dataset? Do I need only voices and its text ? I am completely newbie about TTS.

I prepare a dataset from audio books.Like LJSpeech I have .wav 16 bit PCM Mono sounds. Do I need anything else could you tell me the steps that I should look at.

I checked FAQ but couldnt completly understand steps to preparing own dataset and training model with them. Could you please give me the exact steps?

Check this Docker. It has everything to start training and you can adopt to your needs from there. But TTS is still in active development and you’ll have a lot more to do than for STT.

1 Like