Voice Generation

v1.0

Select menu Window/AI Service Integrations/ElevenLabs Voice Generation to load the window, or right-click an AudioClip field, List, or Array in your project and select ElevenLabs/Voice Generator...

Generating voices via the context menu will populate the field, list or array you selected!

General Settings

The top portion of the window includes general settings and the option to choose "Text to Voice" or "Voice to Voice".

Number of Clips & Concurrent Limit

You can create multiple clips at once. ElevenLabs has a concurrent limit that is determined by your subscription level. If you input a value higher than allowed, an error will occur during generation.

Voice Selection

Voices you have access to are listed here. Click the "Star" icon to toggle a voice as a "Favorite". If there is a preview available, you can listen to that as well.

Voices you've created in ElevenLabs website, including cloned voices, will show in this list as well.

Text to Voice

Text to voice is a quick way to make voice clips for your project.

Text to Synthesize

The content you'd like to hear goes in this field. If you have the Open AI integration, the ChatGPT button can be used to generate or modify the content in the window.

Voice to Voice

Voice to voice allows you to use any existing clip as a source for the AI generated content. You can use an existing clip, or record a new clip directly in the window.

Source Audio

Select the clip you'd like to use as your source here. If you record a clip, it will automatically be populated into this field. Press the "Play" button to listen to the clip without leaving the window.

Recording Audio Path & Name

You can specify the path and name of any recorded audio, making use of variables to better organize your content. The "Text" variable will utilize a transcription of the recording even if Save Transcription to Text File is off.

To record audio, click the "Start Recording" button, and then the "Stop Recording" to end. The recorded clip will be saved and added to the Source Audio field.

Save Transcription to Text File

When toggled on, a .txt file will be created with the transcription of the clip, in the same directory as the clip and with the same name.

Voice Settings

The Text to Voice and Voice to Voice modes share most of the Voice Settings options, but each have some unique options as well.

Text to Voice options

Voice to Voice options

Model

Select the ElevenLabs model you'd like to use.

Stability

Higher values will produce more consistent, but less expressive outputs.

Similarity Boost

Preserves the similarity to the original speaker. Higher values may degrade quality.

Speaker Boost

When true, similarity to the original speaker is increased, at the cost of additional processing time.

Speed (Text to Voice)

Higher values will result in faster speech.

Style Exaggeration (Text to Voice)

Amplifies expressiveness, producing more dramatic results at higher values.

Remove Background Noise (Voice to Voice)

When true, the system will attempt to remove background noise from the source clip.

Use Seed

When true, you can use a specific seed to get consistent results.

Export Options

You can set the Root Folder for your save location, along with the Naming Pattern. Utilize the variables to customize the path and name to organize your files as you'd like.