Coming Soon: Automatic Audio/Video Transcription Powered by Whisper

22nd July 2025

We're excited to announce an upcoming feature for ResourceSpace that brings the power of speech recognition to your audio and video assets — with no third-party dependencies and no data ever leaving your server.

Our new Whisper plugin integrates the highly regarded open-source speech-to-text model developed by OpenAI, enabling ResourceSpace to transcribe spoken content directly from audio and video files and automatically generate subtitles, that can then be downloaded or used by ResourceSpace itself when playing your content.

Whether you're managing interviews, lectures, webinars, podcasts or raw field recordings, you'll soon be able to extract accurate, high-quality transcripts and subtitles — all within your existing ResourceSpace environment.

What Is Whisper?

Whisper is a state-of-the-art automatic speech recognition (ASR) system released by OpenAI. Unlike most transcription tools, Whisper runs entirely locally on our servers (or yours, if you host ResourceSpace yourself).

That means:

• No external services or APIs — nothing is sent to OpenAI

• Full control over your data, even in secure or air-gapped environments

• GDPR compliance by design

If you're handling confidential interviews, sensitive research material, or legally protected recordings, Whisper gives you enterprise-grade speech processing with the privacy and control ResourceSpace is known for.

Seamless Integration with ResourceSpace

Once installed, the Whisper plugin will:

1. Convert uploaded media to audio (WAV)

2. Transcribe speech using Whisper

3. Store the transcription in a designated metadata field

4. Optionally attach subtitle files (.srt) and transcript files (.txt) as alternative downloads

All of this happens automatically in the background.

To assist with the conversion to text, a system wide prompt can be entered which is useful to provide important context, and can contain for example your organisation and product names. This guides the Whisper model and improves accuracy.

Combine with OpenAI GPT for Metadata Magic

If you're already using ResourceSpace's GPT-powered metadata generation, this integration opens up powerful new workflows:

• GPT can now take automatically transcribed audio as input

• Automatically generate titles, descriptions, tags, summaries, or even translations — all from the spoken content

• Enrich your audio and video assets as effortlessly as you do with images and documents

In other words: your videos can now describe themselves.

Coming Soon

The Whisper plugin is currently in final testing and will be available to all customers shortly. As always, it’s open source, secure, and easy to deploy.

Stay tuned for the release announcement — and if you'd like to get early access, just get in touch.

Have questions or want help preparing your server for Whisper? Check out the full setup guide in our Knowledge Base or contact your Customer Success team member for assistance.