Home

Welcome to the Bark-GUI Enhanced wiki!

Commandline Arguments:

• -autolaunch Automatically open Browser with Bark-Tab
• -smallmodels Use small models, for GPUs with less than 10Gb Vram or to speed up process
• -forcecpu Force processing on CPU, if your GPU isn't up to the task
• -enablemps Support for Apple MPS
• -offloadcpu Offload models to CPU

On Windows edit the windows_run.bat or StartBark.bat to customize your launch arguments.

Text-to-Speech Tab

Input any text to let Bark create a Speech, use the Dropbox to choose any voice from the prompts folder (also custom ones). Choose 'None' for undefined (usefull for MAN:/WOMAN: prompts). The Quick Generation checkbox creates audio a lot faster but might be more unstable and perhaps not that subtle as this doesn't use finetuning parameters. Checking Use last generation as history saves voices of each audio chunk to the outputs folder. If you want to use them for output, just copy them into the assets/prompts folder. Contrary to the original Bark, you can input any text length. The result will be created in chunks and merged into 1 audio-file at the end. This can be played from the UI and the WAV-File(s) are saved into the Outputs folder.

A new experimental feature is SSML input and conversion. If you're writing a dialogue like this:

I'm refreshing the Bark Github Page like crazy, there might be a new update every second!

You're just very disturbed

and you've selected Voice 'en_speaker_0' in the dropdown box, clicking on Convert text to SSML the text would be converted to

&lt?xml version="1.0"?&gt
&ltspeak version="1.0" xmlns="http://www.w3.org/2001/10/synthesis"
         xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
         xsi:schemaLocation="http://www.w3.org/2001/10/synthesis
                   http://www.w3.org/TR/speech-synthesis/synthesis.xsd"
         xml:lang="en-US"&gt
         
&ltvoice name="en_speaker_0"&gtI'm refreshing the Bark Github Page like crazy, there might be a new update every second!&lt/voice&gt
&ltvoice name="en_speaker_0"&gtYou're just very disturbed &lt/voice&gt
&lt/speak&gt

As you can see every new line will be split as a potential new line for another voice. Afterwards you could change one of the speaker names (name="en_speaker_0) to create dialogues with fixed voices. If the number of words exceeds the max for Bark generation, new lines will be created reusing the voice from the sentence before that.

Clicking on the Generate Button will automatically detect if this is SSML input or just plain text.

Swap Voice Tab

WIP

Clone Voice Tab

WIP

Settings Tab

WIP