Merge pull request #27 from cbsiamlg/feature/AMLG-7744-update-faster-…

…whisper-and-autoscaling

update faster-whisper and autoscaling

README.md

@@ -10,8 +10,8 @@ This project is a modification of an open source approach to containerizing the
 
 AMLG updated release (v1.2.0) supports following whisper models:
 
-- [openai/whisper](https://github.com/openai/whisper)@[v20230124](https://github.com/openai/whisper/releases/tag/v20230124)
-- [faster-whisper](https://github.com/guillaumekln/faster-whisper)@[0.6.0](https://github.com/guillaumekln/faster-whisper/releases/tag/v0.6.0)
+- [openai/whisper](https://github.com/openai/whisper)@[v20231117](https://github.com/openai/whisper/releases/tag/v20231117)
+- [faster-whisper](https://github.com/guillaumekln/faster-whisper)@[1.0.3](https://github.com/guillaumekln/faster-whisper/releases/tag/v1.0.3)
 
 ## Usage
 
@@ -214,4 +214,4 @@ docker run -d -p 9000:9000 -e ASR_MODEL=large -v //c/tmp/whisper:/root/.cache/wh
 
 ## Updates to the models
 
-Going forward the poetry updates to whisper and faster-whisper will be run on a remote machine. The updated models will be stored in the GAR per the usual process. The docker images will be updated with the new models and pushed to the GAR. The images will be tagged with the model version. The latest tag will be updated to point to the latest model version. The cpu-latest is built and pushed as an AMD64 image and tagged as `cpu-latest` in the dev GAR only. You may not be able to build locally unless we add more logic for 
+Going forward the poetry updates to whisper and faster-whisper will be run on a remote machine. The updated models will be stored in the GAR per the usual process. The docker images will be updated with the new models and pushed to the GAR. The images will be tagged with the model version. The latest tag will be updated to point to the latest model version. The cpu-latest is built and pushed as an AMD64 image and tagged as `cpu-latest` in the dev GAR only. You may not be able to build locally unless we add more logic for 

infrastructure/whisper/values.prod.yaml

@@ -22,8 +22,8 @@ whisperService:
 whisperHPA:
   autoscale:
       min: 3
-      max: 15
+      max: 75
       memoryUtilization: 75
 
 ingress:
-  host: whisper.amlg.io
+  host: whisper.amlg.io

infrastructure/whisper/values.yaml

@@ -22,8 +22,8 @@ whisperService:
 whisperHPA:
   autoscale:
       min: 3
-      max: 6
+      max: 26
       memoryUtilization: 75
 
 ingress:
-  host: devwhisper.amlg.io
+  host: devwhisper.amlg.io