PDF to Podcast Deployment Guide

PDF to Podcast Deployment Guide#

Solution Blueprints are provided as Helm Charts. The recommended approach to deploy them is to pipe the output of helm template to kubectl apply -f -. We do not recommend helm install, which by default uses a Secret to keep track of the related resources. This does not work well with Enterprise clusters that often have limitations on the kinds of resources that regular users are allowed to create.

This blueprint supports AMD Instinct (default) and AMD Radeon platforms. Unless otherwise specified, the commands below cover the default Instinct deployment. For deployment with Radeon, see:

Multi-platform Support#

The chart ships defaults for two platforms, selected with --set global.platform=<platform>: instinct (GPU, the default) and radeon (GPU). Each selects a matching AIM image and resource profile for the LLM; inspect them with helm show values . --jsonpath '{.llm.platformDefaults}'.

Helm note: Built and tested on Helm 3.17 or higher. On Helm v4, if the piped kubectl apply is rejected, run helm pull oci://registry-1.docker.io/amdenterpriseai/aimsb-pdf-to-podcast --untar first and template the local ./aimsb-pdf-to-podcast directory instead.

Both the LLM and TTS services are deployed automatically via subchart dependencies (aimchart-llm and aimchart-qwen-tts). The environment variables APP_LLM_URL and APP_TTS_BASE_URL are auto-configured from the subchart service URLs.

AMD Instinct (GPU, default)#

To deploy the blueprint, run the following command:

name="my-deployment"
namespace="my-namespace"
helm template $name oci://registry-1.docker.io/amdenterpriseai/aimsb-pdf-to-podcast \
  | kubectl apply -f - -n $namespace

AMD Radeon (GPU)#

To deploy the blueprint, run the following command:

name="my-deployment"
namespace="my-namespace"
helm template $name oci://registry-1.docker.io/amdenterpriseai/aimsb-pdf-to-podcast \
  --set global.platform=radeon \
  | kubectl apply -f - -n $namespace

Using an existing deployment or external LLM#

By default, any required AIMs are deployed by the helm chart. If you already have a compatible AIM deployed, you can use that instead, and reuse resources.

To use an existing deployment or external LLM, set the value llm.existingService to that endpoint. Then, any other values you pass in the llm mapping are simply ignored, and your existing service is used instead. You should use the Kubernetes Service name, or if the service is in a different namespace, you can use the long form <SERVICENAME>.<NAMESPACE>.svc.cluster.local:<SERVICEPORT>. If needed, you can pass a whole URL.

Full example command:

name="my-deployment"
namespace="my-namespace"
servicename="aim-llm-my-model-123456"
helm template $name oci://registry-1.docker.io/amdenterpriseai/aimsb-pdf-to-podcast \
  --set llm.existingService=$servicename \
  | kubectl apply -f - -n $namespace

API Key and Model Configuration for External LLM#

When using an external LLM service, you can optionally configure API authentication and model override:

  • llm.apiKey (optional): Bearer token for API authentication

  • llm.model (optional): Explicit model id to use

name="my-deployment"
namespace="my-namespace"
api_url="https://llm-api.example.com"
api_key="<YOUR_API_KEY>"
model_name="openai/gpt-oss-20b"

helm template $name oci://registry-1.docker.io/amdenterpriseai/aimsb-pdf-to-podcast \
  --set llm.existingService=$api_url \
  --set llm.apiKey=$api_key \
  --set llm.model=$model_name \
  | kubectl apply -f - -n $namespace

Using Kubernetes Secrets for API Key#

name="my-deployment"
namespace="my-namespace"
secretname="my-secretname"
api_url="https://llm-api.example.com"

kubectl create secret generic $secretname -n $namespace \
  --from-literal=api-key="<YOUR_API_KEY>"

helm template $name oci://registry-1.docker.io/amdenterpriseai/aimsb-pdf-to-podcast \
  --set llm.existingService=$api_url \
  --set llm.apiKeySecretRef.name=$secretname \
  --set llm.apiKeySecretRef.key=api-key \
  | kubectl apply -f - -n $namespace

Using an existing deployment or external TTS#

Similarly, to use an existing TTS service instead of the default aimchart-qwen-tts deployment, set qwen-tts.existingService to the endpoint. The TTS service must expose an OpenAI-compatible API (POST /v1/audio/speech).

name="my-deployment"
namespace="my-namespace"
ttsservice="my-tts-service"
helm template $name oci://registry-1.docker.io/amdenterpriseai/aimsb-pdf-to-podcast \
  --set qwen-tts.existingService=$ttsservice \
  | kubectl apply -f - -n $namespace

If the TTS service requires an API key, pass it via:

  --set pythonServices.app.env.APP_TTS_API_KEY="<your_tts_api_key>"

Default AIM image and GPU compatibility#

By default, the chart deploys Meta Llama 3.3 70B with this AIM: amdenterpriseai/aim-meta-llama-llama-3-3-70b-instruct:0.11.1

On newer GPUs, this default image may not be the best match and can fail to start or run sub-optimally. To choose a newer AIM or deploy a different LLM, override llm.image to a compatible image. See the catalog of available AIMs for options.

Example:

name="my-deployment"
namespace="my-namespace"
helm template $name oci://registry-1.docker.io/amdenterpriseai/aimsb-pdf-to-podcast \
  --set llm.image=amdenterpriseai/aim-meta-llama-llama-3-3-70b-instruct:<NEWER_TAG> \
  | kubectl apply -f - -n $namespace

Using a custom image and imagePullSecrets#

To use a custom version of the image from another repo, you can set the image and imagePullSecrets, if necessary, from the command line as follows:

imagerepository="my-repo/custom-pdf-to-podcast-application"
imagetag="0.0.1"
uiimagerepository="my-repo/custom-pdf-to-podcast-ui"
uiimagetag="0.0.1"
secretname="my-registry-secret"
helm template $name oci://registry-1.docker.io/amdenterpriseai/aimsb-pdf-to-podcast \
  --set image.repository=$imagerepository \
  --set image.tag=$imagetag \
  --set uiImage.repository=$uiimagerepository \
  --set uiImage.tag=$uiimagetag\
  --set imagePullSecrets[0].name=$secretname \
  | kubectl apply -f - -n $namespace

Configuration Highlights#

  • LLM: set llm.existingService to reuse an external LLM; otherwise the chart deploys the default via aimchart-llm dependency.

  • TTS: set qwen-tts.existingService to reuse an external OpenAI-compatible TTS service; otherwise the chart deploys the default via aimchart-qwen-tts dependency. APP_TTS_BASE_URL is auto-configured from the subchart.

  • Secrets: configure API and TTS keys under pythonServices.app.env:

    • APP_TTS_API_KEY - API key for the TTS service (if required by the TTS backend)

    • APP_API_KEY - Optional API key for authentication

    • APP_CELERY_BROKER_URL and APP_CELERY_BACKEND_URL - Auto-configured if not set

  • Storage: adjust persistent volume settings in values.yaml:

    • storage.ephemeral - Temporary storage (20Gi default)

    • storage.appStorage - Shared storage for PDFs (10Gi default, ReadWriteMany)

    • storage.dshm - Shared memory size (32Gi default)

  • Resources: adjust CPU and memory limits in resources section (default: 500m CPU requests, 4 CPU limits, 4Gi memory requests, 8Gi memory limits).

Connecting#

Option 1: Port Forwarding#

To connect to the UI, port-forward port 7860. The UI will then be available at http://localhost:7860.

kubectl port-forward svc/aimsb-pdf-to-podcast-$name-frontend 7860:7860 -n $namespace

Option 2: HTTPRoute (Gateway Access)#

If your cluster has a Gateway API compatible gateway (e.g., Kubernetes Gateway, Istio, etc.), you can enable HTTPRoute creation to route traffic through the gateway.

Prerequisites:

  • A Gateway named https must exist in the envoy-gateway-system namespace (or configure a different gateway).

  • The Gateway must be properly configured with listeners.

Enabling HTTPRoute:

Use --set http_route.enabled=true in the helm template command to enable HTTPRoute creation:

helm template $name oci://registry-1.docker.io/amdenterpriseai/aimsb-pdf-to-podcast \
    --set http_route.enabled=true \
     | kubectl apply -f - -n $namespace

Obtaining the URL:

The URL to access the blueprint via HTTPRoute is formed by the service name and the hostname of the gateway. Use this command to produce the URL by querying the hostname from the cluster:

echo "https://aimsb-pdf-to-podcast-$name$(kubectl get gtw -A -o jsonpath='{.items[*].spec.listeners[?(@.name=="https")].hostname}' | tr -d \*)/"

Clean Up#

When you are finished, remove the deployed resources:

helm template $name oci://registry-1.docker.io/amdenterpriseai/aimsb-pdf-to-podcast \
  | kubectl delete -f - -n $namespace
Contents