Deepgram released a new version of its on-premises solution.
On-Premises Release 220831 Docker Hub Images
deepgram/onprem-api:1.64.0
deepgram/onprem-engine:3.33.0
Minimum required NVIDIA driver version: >=450.80.02
deepgram/onprem-license-proxy:1.2.1
deepgram/onprem-billing:1.4.0
deepgram/onprem-metrics-server:2.0.0
This Release Contains The Following Changes
Support for improved Diarization. The new Diarization model should be placed into the model_manager search path and can be invoked by using the diarize=true parameter in the ASR request. Please visit the Diarization documentation page to learn more.
Improved Diarization beta users should contact Deepgram support to receive the latest version of the new Diarization model, after which the beta [[features.diarizer]] section in engine.toml can be removed.
ASR requests can invoke a specific model by passing the the model UUID instead of the model name for the model URI parameter.
Streaming Callback requests can now receive the transcript response in the requesting WebSocket (in addition to the callback) by including the inspect_response=true parameter in the ASR request.
Support for the deprecated weights field in engine.toml has been removed. Use the path field moving forward.
Streaming audio which is received >1.25x real-time speed is now throttled.
Resolves a stability issue in the License Proxy Server.