Attention:

This is a very beta feature that is only available upon request.

Open the OBS Remote

Under the Audio Mixer, right click the audio levels for the audio source you'd like to transcribe (eg Ingest Player) and select Filters

Under Audio/Video Filters add the LocalVocal Transcription filter.

Adding this filter for the first time will add a Text (FreeType 2) source named LocalVocal Subtitles to the current scene.
- You can add more Text (FreeType 2) sources as needed as transcription targets.
- Support for Text (Pango) source may be available in the future.

Configuration

Output Destination can be None or any Text (FreeType 2) source.

Choose your Transcription model. Whisper Tiny, Small, Medium have been tested with success.

HuggingFace and other models have not yet been tested.

Cloud Translation

Attention:

Do not use Local Translation

Select the Cloud Translation checkbox.

Currently Google Cloud Translation and DeepL are supported but you will have to provide your own translation API key from the respective service.

Both offer 500,000 characters of free translation per month but require registering and providing billing information to receive an access key.

More details for pricing and limits: Google / DeepL

Please read the terms carefully.

IRLToolkit will not be liable for charges incurred from API overusage.

Create a Google Cloud Project:
- Go to the Google Cloud Console.
- Click on Select a project at the top and choose New Project.
- Give your project a name and click Create.
Enable the Translation API:
- In the Google Cloud Console, navigate to APIs & Services → Library.
- Search for Cloud Translation API.
- Click on it and select Enable.
Set Up Billing:
- Google requires billing to use the API, even for free tier usage.
- Go to Billing in the Console and link a billing account to your project.
Create API Credentials:
- Go to APIs & Services → Credentials.
- Click + Create Credentials → API Key.
- Your API key will be generated and displayed.

Paste the API key into the Access Key field of obs-localvocal.

Create a DeepL Account:
- Go to the DeepL website.
- Sign up for a DeepL API Free or a Pro plan, depending on your needs.
Access Your API Key:
- After subscribing, go to your DeepL Account page.
- Under the API Keys & Limits section, you'll find your API Key.

Paste the API key into the Access Key field of obs-localvocal.

You are welcome to test the closed captioning output, but it may get more and more behind in some situations.

Outputting to a text source seems more responsive, at the moment.

Changing to Advanced mode enables access to word filtering for words that the AI may hallucinate.

Other settings are outside the scope of this guide.