Is this page helpful?

End-to-End Demo Chart Configuration#

This page covers Helm values for the sender and receiver components of the nvidia-studio-voice-h4m-sample chart. NIM service keys within the demo chart live under nvidia-studio-voice-h4m-service.<key>. For more information, refer to Common Helm Configuration and Pipeline Configuration.

Sender#

Configuration

Helm Key (sender.*)

Comment

Enable sender

sender.enabled

Allowed: Boolean; two-phase bring-up (receiver and NIM service first, then sender) recommended.

Naming / scale

sender.appName, sender.replicas

—

Node

sender.nodeSelector.hostname

—

Image

sender.image.repository, sender.image.tag, sender.image.secret

—

Input asset

sender.inputAssets.audioFile

TS file path inside sender container.

Custom input PVC

sender.inputData.pvc.enabled, sender.inputData.pvc.claimName

Mount custom audio at /workspace/assets.

SMPTE ST 2110 ports

sender.st2110.multicastIp, sender.st2110.audioPort, sender.st2110.multicastTtl

Site-specific.

NMOS

sender.nmos.*

Same semantics as NIM service nmos.*; refer to Common Helm Configuration.

Input Asset Requirements#

The sender expects an MPEG Transport Stream (.ts) file with the following specifications:

  • Audio stream: Opus, 48 kHz, mono.

The sender container includes a bundled sample file at /workspace/assets/studio_voice_48k_2_looped_10min.ts (48 kHz, mono, 10 minutes looped).

Receiver#

Configuration

Helm Key (receiver.*)

Comment

Enable receiver

receiver.enabled

—

Naming / scale

receiver.appName, receiver.replicas

—

Node

receiver.nodeSelector.hostname

—

Image

receiver.image.*

—

Receive mode

receiver.receiverMode

Default: processed.

SMPTE ST 2110

receiver.st2110.multicastIp, receiver.st2110.audioPort

Must match NIM output.

Audio encoding

receiver.audioParams.opusBitrate

Default: 128000.

NMOS

receiver.nmos.*

—

SRT output

receiver.srtPort.internal, receiver.srtPort.external

Defaults: internal 9090, external 32090.

See Also#