Need help downloading NVIDIA Maxine Audio Effects SDK for real-time speech processing project

AI & Data Science Deep Learning (Training & Inference) Maxine
,
1

Hello,

I am a developer working on a real-time subtitle generation system and I would like to integrate NVIDIA Maxine Audio Effects SDK into my pipeline.

My system currently works as follows:

Audio pipeline:

Video / media source
→ Voicemeeter audio routing
→ NVIDIA Broadcast (noise removal)
→ Whisper large-v3 speech recognition
→ DeepL EN→FI translation
→ Finnish grammar and compound word correction
→ live subtitles overlay

The system runs locally on Windows 11 with an NVIDIA RTX 3060 (12 GB) GPU.

I am now trying to improve the audio quality before speech recognition by adding NVIDIA Maxine Audio Effects SDK (denoiser / dereverb / speech enhancement).

However I am having difficulty finding the correct downloads for:

• Maxine Audio Effects SDK (Windows developer package)
• the required AI models (denoiser / dereverb)
• the correct download links or NGC registry packages

Could someone please clarify:

  1. Which exact Maxine SDK package should be used for real-time speech enhancement?

  2. Where can the latest Windows developer package be downloaded?

  3. Which NGC model packages are required for the audio effects?

  4. Are there any license or developer program requirements to download the SDK?

  5. Is the SDK still actively maintained or replaced by NVIDIA Broadcast / Maxine services?

My goal is to integrate Maxine as an additional preprocessing layer before Whisper in a Python + Rust audio pipeline.

Any help, documentation links, or download instructions would be greatly appreciated.

Thank you!

2

Hi Aurora,

Here are some helpful resources that I dug up:

Hope that helps!

Thanks,

Aharpster

3

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.