Testing

[Smartking1]

I have successfully run the colab notebook and saved the speaker.npz file. The issue now is how do I test it on a new data

[gitmylo]

What do you mean by that?

[Smartking1]

how do I test the clone on a text data

[gitmylo]

Use bark to generate it and use the npz as the speaker file.

An easy gui which lets you do this for example: https://github.com/gitmylo/audio-webui But you could also just set the speaker file in the official bark repository.

[Smartking1]

Thanks Is there a sample code to generate it with bark or a guide to do that

[gitmylo]

under this section of the bark readme: https://github.com/suno-ai/bark/tree/main#-voice-presets

[gitmylo]

instead of putting a preset speaker name, you can put the path to your npz file

[Smartking1]

Thanks so much I have also launched the web_ui interface on Gradio

Which of the options can I use for voice cloning

[gitmylo]

on the text to speech tab, you can select bark

there's also rvc speech conversion, which takes audio as the input, and replaces the voice, but you need to train that one.

[Smartking1]

Alright Working now