Skip to navigation

Overview

The nvidia-studio-voice-h4m-sample Helm chart deploys a complete Studio Voice demo pipeline on Holoscan for Media. This includes:

  • Sender pipeline.
  • Studio Voice NIM service (nvidia-studio-voice-h4m-service).
  • Receiver pipeline (SMPTE ST 2110 or NMOS).
  • SRT output for preview.

The Studio Voice NIM processes incoming audio streams and performs real-time AI-based speech enhancement (noise suppression, dereverberation, restoration of clarity). Enhanced audio is transmitted via ST 2110-30.

Installation

Prerequisites

Complete all prerequisite steps (Rivermax license, image pull and model pull secrets, high-speed network attachment) before running helm install. For details, refer to Getting Started.

Pull the Chart

If not already done, add the Helm repository and pull the chart:

helm pull nim-repo/nvidia-studio-voice-h4m-sample --version 1.2.0

For the full repository setup, refer to Pull Helm Charts.

Helm Installation

The chart includes a default values.yaml. For SMPTE ST 2110 static or NMOS pipelines, copy the configuration from Starter Configuration Files into values-st2110.yaml or values-nmos.yaml, adjust fields, and run the matching install command.

The default values.yaml ships with example values for cluster-specific fields such as node selector (example-gpu-node), image pull secret, network name, and scheduler. Before deploying, you must update these fields to match your cluster. The easiest approach is to use the global overrides so that each value is specified once for all components (sender, receiver, and NIM service). For details, refer to Global Overrides.

Minimal deployment using global overrides (NMOS mode):

helm upgrade --install studio-voice-h4m-sample \
nvidia-studio-voice-h4m-sample-1.2.0.tgz \
--set global.nodeSelector.hostname=<gpu-node-name> \
--set global.image.secret=<image-pull-secret> \
--set global.network.name=<multus-net-attach-def> \
--set global.schedulerName=<scheduler-name> \
--set nvidia-studio-voice-h4m-service.ngc.secretName=<model-pull-secret>

Alternatively, set values per component when the sender, receiver, and NIM service run on different nodes or use different pull secrets:

helm upgrade --install studio-voice-h4m-sample \
nvidia-studio-voice-h4m-sample-1.2.0.tgz \
--set sender.nodeSelector.hostname=<gpu-node-name> \
--set receiver.nodeSelector.hostname=<gpu-node-name> \
--set nvidia-studio-voice-h4m-service.nodeSelector.hostname=<gpu-node-name> \
--set sender.image.secret=<image-pull-secret> \
--set receiver.image.secret=<image-pull-secret> \
--set nvidia-studio-voice-h4m-service.image.secret=<image-pull-secret> \
--set nvidia-studio-voice-h4m-service.ngc.secretName=<model-pull-secret>

For all available Helm values, refer to the Configuration Reference.

First, deploy the receiver and NIM service. When the pipeline is ready, enable the sender. This procedure prevents audio from accumulating in receiver queues before the NIM is ready.

# Phase 1: receiver + NIM service only
helm upgrade --install studio-voice-h4m-sample \
nvidia-studio-voice-h4m-sample-1.2.0.tgz \
--set sender.enabled=false
kubectl rollout status deployment/<receiver-appName> --timeout=180s
kubectl rollout status deployment/<nim-appName> --timeout=300s
# Phase 2: enable sender
helm upgrade --install studio-voice-h4m-sample \
nvidia-studio-voice-h4m-sample-1.2.0.tgz
kubectl rollout status deployment/<sender-appName> --timeout=180s

On Red Hat OpenShift, replace kubectl with oc.

A rollout status timeout is not a failure; it means the pod did not become ready within the allotted time. The NIM pod might still be pulling its image or initializing. If a timeout occurs, check the pod state before taking action:

kubectl get pods -o wide
kubectl describe pod <pod-name>

Look for Pulling image in the events (normal; wait longer) or CrashLoopBackOff / ErrImagePull (actionable: check secrets and node resources).

After installation, verify all pods are running:

kubectl get pods

Confirm that the sender, NIM service, and receiver pods all show READY 1/1 before continuing.


Sender Sample Media

Input Media Files

The sender expects an MPEG Transport Stream (.ts) file containing an audio stream for Studio Voice enhancement.

Codecs

  • Audio codec: Opus, 48 kHz, mono.

Bundled Sample Files

The sender container includes a sample .ts file:

/workspace/assets/studio_voice_48k_2_looped_10min.ts

FileSample RateChannels
studio_voice_48k_2_looped_10min.ts48 kHzMono

Using a Custom Input Asset

To test with your own audio, upload a .ts file to a PersistentVolumeClaim (PVC) and enable the built-in PVC mount on the sender:

sender:
inputAssets:
audioFile: /workspace/assets/my-custom-audio.ts
inputData:
pvc:
enabled: true
claimName: studiovoice-sender-input-data

The PVC is mounted at /workspace/assets inside the sender pod, overlaying the default bundled assets.


Starter Configuration Files

Copy the appropriate configuration into values-st2110.yaml or values-nmos.yaml, adjust fields for your environment, and pass it to helm upgrade --install with -f.

For a full reference of all Helm keys and pipeline tuning parameters, refer to the Configuration Reference.

SMPTE ST 2110 Configuration — Static (values-st2110.yaml)

sender:
enabled: true
appName: nvidia-studio-voice-sender-st2110
replicas: 1
nodeSelector:
hostname: example-gpu-node
image:
repository: nvcr.io/nim/nvidia/studio-voice-h4m-sample-client
tag: "1.2.0"
secret: ngc-image-secret
network:
name: media-a-tx-net
inputAssets:
audioFile: /workspace/assets/studio_voice_48k_2_looped_10min.ts
st2110:
multicastIp: "234.5.8.9"
audioPort: 5003
multicastTtl: 24
nmos:
enabled: false
label: Studio Voice Audio Sender
receiver:
enabled: true
appName: nvidia-studio-voice-receiver-st2110
replicas: 1
nodeSelector:
hostname: example-gpu-node
image:
repository: nvcr.io/nim/nvidia/studio-voice-h4m-sample-client
tag: "1.2.0"
secret: ngc-image-secret
network:
name: media-a-tx-net
receiverMode: processed
st2110:
multicastIp: "234.5.8.9"
audioPort: 5004
multicastTtl: 24
audioParams:
opusBitrate: 128000
nmos:
enabled: false
label: Studio Voice Audio Receiver
srtPort:
internal: 9090
external: 32090
nvidia-studio-voice-h4m-service:
enabled: true
appName: nvidia-studio-voice-nim-st2110
replicas: 1
nodeSelector:
hostname: example-gpu-node
image:
repository: nvcr.io/nim/nvidia/studio-voice-h4m
tag: "1.2.0"
pullPolicy: IfNotPresent
secret: ngc-image-secret
network:
name: media-a-tx-net
nmos:
enabled: false
httpPort: 9010
seed: nvidia-studio-voice-nim-st2110
audio:
pcmFormat: S24BE
samplingRate: 48000
numChannels: 1
input:
audio:
sessionName: audio_in
localInterfaceName: net1
hostIp: "234.5.8.9"
hostPort: 5003
hostNumSubnetBits: 24
output:
audio:
sessionName: audio_out
localInterfaceName: net1
hostIp: "234.5.8.9"
hostPort: 5004
hostNumSubnetBits: 24
studiovoice:
nmosEnabled: false
nimEnabled: true

NMOS Configuration (values-nmos.yaml)

sender:
enabled: true
appName: nvidia-studio-voice-sender-nmos
replicas: 1
nodeSelector:
hostname: example-gpu-node
image:
repository: nvcr.io/nim/nvidia/studio-voice-h4m-sample-client
tag: "1.2.0"
secret: ngc-image-secret
network:
name: media-a-tx-net
inputAssets:
audioFile: /workspace/assets/studio_voice_48k_2_looped_10min.ts
nmos:
enabled: true
hostname: studio-voice-sender.local
description: Studio Voice Audio Sender
label: Studio Voice Audio Sender
receiver:
enabled: true
appName: nvidia-studio-voice-receiver-nmos
replicas: 1
nodeSelector:
hostname: example-gpu-node
image:
repository: nvcr.io/nim/nvidia/studio-voice-h4m-sample-client
tag: "1.2.0"
secret: ngc-image-secret
network:
name: media-a-tx-net
receiverMode: processed
nmos:
enabled: true
hostname: studio-voice-receiver.local
description: Studio Voice Audio Receiver
label: Studio Voice Audio Receiver
srtPort:
internal: 9090
external: 32090
nvidia-studio-voice-h4m-service:
enabled: true
appName: nvidia-studio-voice-nim-nmos
replicas: 1
nodeSelector:
hostname: example-gpu-node
image:
repository: nvcr.io/nim/nvidia/studio-voice-h4m
tag: "1.2.0"
pullPolicy: IfNotPresent
secret: ngc-image-secret
network:
name: media-a-tx-net
nmos:
enabled: true
httpPort: 9010
seed: nvidia-studio-voice-nim-nmos
hostname: studio-voice-nim.local
description: Studio-Voice-NIM
label: Studio-Voice-NIM
audio:
pcmFormat: S24BE
samplingRate: 48000
numChannels: 1
input:
audio:
sessionName: audio_in
output:
audio:
sessionName: audio_out
studiovoice:
nmosEnabled: true
nimEnabled: true

In NMOS mode, the transport parameters (multicast IP address, port, subnet) are negotiated dynamically by the NMOS Connection Manager over IS-04 and IS-05, so the preceding NIM service block needs only sessionName for each input/output. The values are discovered at runtime.

On Holoscan for Media clusters, the NMOS Connection Manager web UI is typically reached from a browser on the cluster network. For remote graphical access, refer to Chrome Remote Desktop in Getting Started.

In NMOS mode, connect receivers to the Studio Voice NIM before connecting the sender via the NMOS Connection Manager UI. In SMPTE ST 2110 mode (static), ensure that the “sender → NIM → receiver IP address and port” chain is consistent across all values.

Install

SMPTE ST 2110 (static):

helm upgrade --install studio-voice-h4m-sample \
nvidia-studio-voice-h4m-sample-1.2.0.tgz \
-f values-st2110.yaml

NMOS is enabled by default. To deploy in fixed SMPTE ST 2110 mode (static), append the following:

--set sender.nmos.enabled=false \
--set receiver.nmos.enabled=false \
--set nvidia-studio-voice-h4m-service.nmos.enabled=false

NMOS:

helm upgrade --install studio-voice-h4m-sample \
nvidia-studio-voice-h4m-sample-1.2.0.tgz \
-f values-nmos.yaml

View Output via SRT

Get the internal IP addressof the node:

kubectl get nodes -o wide

The SRT output is on port 32090 by default. Open the stream in VLC or ffplay:

ffplay "srt://<internal-ip>:32090"

Disabling Components

Individual pipeline components can be disabled during helm install or helm upgrade without removing the release:

--set sender.enabled=false
--set receiver.enabled=false
--set nvidia-studio-voice-h4m-service.enabled=false

Only the selected component stops; the rest of the pipeline continues to run.


Verify

helm status studio-voice-h4m-sample
kubectl get pods -o wide

Verify configured Helm values:

helm get values studio-voice-h4m-sample

On Red Hat OpenShift, replace kubectl with oc. For log access, refer to Observability.


Uninstall

helm uninstall studio-voice-h4m-sample

The *-model-cache and *-nim-logs PVCs carry helm.sh/resource-policy: keep and are not removed by helm uninstall. This preserves the cached model across reinstalls; clean them up manually when no longer needed:

kubectl delete pvc -l app.kubernetes.io/instance=studio-voice-h4m-sample -n <your-namespace>

End-to-End Verification

  1. Create values-st2110.yaml or values-nmos.yaml from Starter Configuration Files and run the matching Install command.
  2. Confirm all pods show READY 1/1 with kubectl get pods.
  3. Open the SRT preview; refer to View Output via SRT. For VLC steps, NMOS wiring, and screenshots, refer to Verification.
  4. Run helm uninstall studio-voice-h4m-sample.

For troubleshooting, refer to Advanced Usage.