Whisper ASR webservice
For faster inference, switch the image to onerahmet/openai-whisper-asr-webservice:latest-gpu, set ASRDEVICE=cuda, install the NVIDIA Container Toolkit, and add the same deploy.resources.reservations.devices NVIDIA block used in stacks/ollama/docker-compose.yml.
Standard isolation
What to know
- Access
- Caddy route available; choose an access policy
- Login
- Application login
- Shared services
- No shared resources
- Backups
- Backups recommended
- Resource size
- Small
- Lifecycle
- Stable
Ready for the technical setup?
The stack README documents configuration, storage, networking, deployment, verification, upgrades, and troubleshooting.
View technical setup