Hello David,
Welcome to our community.
Regarding your question related to the WhisperSpeechToText module Convert speech to text to generate live stream captions with OpenAI Whisper | Wowza Docs & APIs
The whisperSpeechToText module doesn’t work with OpenAI’s public Whisper API.
Instead, the module expects to talk to a local server that runs the Whisper engine something like a Docker container or a custom-hosted setup using the Whisper open-source code.
- You’ll need to either use our Docker Compose setup from GitHub (which includes a working Whisper server), GitHub - WowzaMediaSystems/wse-plugin-caption-handlers: Generate captions using ASR. Implementations using Azure Speech Services SDK or Whisper
Or - Host Whisper locally and configure the module to point to that local server. GitHub - openai/whisper: Robust Speech Recognition via Large-Scale Weak Supervision
I hope you find this useful.
Wishing you continued success and growth!
Regards, Umair