Sections

  • Home
  • Archive
  • LLM Prompts
  • Posts

BTC

Bitcoin QR Code

Recently Modified

  • SMTP Test With PowerShell on 2025-07-02
  • Videos as Teams Backgrounds on 2025-07-02
  • UDM Parameters for Google Search on 2025-06-18
  • Troubleshoot Crashing Apps with ProcDump & WinDbg on 2025-05-01
  • Stub Title on 2025-03-07
  • Automated IIS Application Pool Restart with PowerShell on 2024-10-16
  • Managing Microsoft Office Versions with OfficeC2RClient on 2024-09-10
  • Automated Batch Image Compression with Python on 2024-07-30
  • How to Find Your Public IP Address on 2024-06-24
  • Complete DNS Records Reference Guide on 2024-06-06

NVIDIA Maxine Windows Audio Effects SDK

Published: April 20, 2022 | Last Modified: May 13, 2025

Tags: audio sound SDK

Categories: Shell


This is my tentative workflow for cleaning-up poor audio using the NVIDIA Maxine Windows Audio Effects SDK.

Download (requires NVIDIA developer account)

NVIDIA Maxine Documentation

The Audio Effects SDK only accepts audio that is in .wav format sampled at 8000Hz single-channel, 16000Hz single-channel, or 48000Hz single-channel. This can be resampled using ffmpeg.

  • Convert a .mp3 file to 8kHz, single-channel PCM:
ffmpeg -i input.mp3 -ar 8000 -ac 1 8000.wav
  • Convert a .m4a file to 16kHz, single-channel PCM:
ffmpeg -i input.mp3 -ar 16000 -ac 1 16000.wav
  • Convert a .wav file to 48kHz, single-channel PCM:
ffmpeg -i input.mp3 -ar 48000 -ac 1 48000.wav

A config file has to be fed to a batch script. The “effects_demo” includes sample config files for different GPU architectures. I hava an NVIDIA RTX 4080, so I would customize the “denoiser48k_cfg_ada.txt” config file and run it with the “run_denoiser_48k_ada.bat” batch file. Example config:

# Effect.
# Supported values are: denoiser/dereverb/dereverb_denoiser/aec/superres
effect dereverb_denoiser
# Point this to the model file.
# This indicates 48k model for denoiser effect for ADA supported GPU architecture is picked from models folder
# Similarly, this path can be modified as per user's choice of effect and sample rate (8k/16k/48k depending on effect)
model models\ada\dereverb_denoiser_48k.trtpkg
# Noisy input file
# 48k Input file is picked from denoiser folder. 
# User can modify below line to pick their own file as input.
input_wav input_files\48000.wav
# Denoised audio data will be saved to this file.
# Output can be dumped at user specifid location too. In this case, Output will be saved to current folder.
output_wav 48000.wav
# Set to 1 for real time mode i.e. audio data will be processed 
# at same speed like that of an audio input device like
# microphone. Since the denoising is faster that real time, the
# processing will be equal to audio file duration.
real_time 0
# Intensity Ratio
intensity_ratio 0.5
# Enable VAD
enable_vad 1

Categories

  • Active Directory (4)
  • AI (3)
  • Azure AD (1)
  • C# (2)
  • C++ (1)
  • Computer Vision (1)
  • DNS (1)
  • Exchange (2)
  • Google (1)
  • Image Processing (2)
  • Java (32)
  • JavaScript (17)
  • Machine Learning (3)
  • MASM (3)
  • Media Processing (1)
  • Microsoft 365 (2)
  • Microsoft Office (1)
  • Microsoft Teams (1)
  • Networking (4)
  • Nodejs (1)
  • Office 365 (1)
  • P5.js (9)
  • PowerShell (25)
  • Processing (14)
  • Programming (1)
  • Python (19)
  • Reference (1)
  • Security (8)
  • Shell (16)
  • Stub (1)
  • System Administration (4)
  • Teams (1)
  • Visualization (1)
  • Web Administration (1)
  • Web Development (2)
  • Windows (9)

Tags

  • 10PRINT (1)
  • 3d-Modeling (1)
  • 3n+1 (1)
  • Account Management (1)
  • Acl (1)
  • Active-Directory (10)
  • Ad Sync (1)
  • Ai (9)
  • Android (1)
  • Animation (10)
  • Api (2)
  • Arrays (1)
  • Assembly (3)
  • Audio (3)
  • Audio Conversion (1)
  • Automation (13)
  • Azure (4)
  • Azure Ad Connect (1)
  • Base64 (1)
  • Bat (2)
  • Batch-Processing (1)
  • Bipartite Graph (1)
  • Bitset (1)
  • Buddhabrot (1)
  • Calendars (1)
  • Channel Management (1)
  • Client-Side (1)
  • Cmd (1)
  • Coding Challenge (15)
  • Collaboration (1)
  • Collatz Conjecture (1)
  • Command-Line (6)
  • Compliance (1)
  • Computer-Vision (3)
  • Coqui-Tts (1)
  • Counting Sort (1)
  • Creative-Coding (1)
  • Cuda (2)
  • Curl (1)
  • Cybersecurity (1)
  • Dag (1)
  • Data-Visualization (5)
  • Debugging (1)
  • Decoding (1)
  • Depth Estimation (1)
  • Device Management (1)
  • Directed Acyclic Graph (1)
  • Directory-Services (1)
  • Disjoint Set (1)
  • Distance (1)
  • Dkim (1)
  • Dmarc (1)
  • Dns (2)
  • Domain (1)
  • Domain Controller (1)
  • Domain Security (1)
  • Domain-Management (1)
  • Download (2)
  • Drivers (1)
  • Drives (1)
  • Education (1)
  • Email (1)
  • Email Management (1)
  • Email Security (2)
  • Email-Archiving (1)
  • Events (1)
  • Exchange (1)
  • Exchange-Management (1)
  • Exchange-Online (2)
  • ExchangeOnlineManagement (3)
  • Ffmpeg (4)
  • Fibonacci (1)
  • File-Permissions (1)
  • File-System (1)
  • Film (1)
  • Filtering (1)
  • Finance (1)
  • Firewall (1)
  • Flask (1)
  • Fractal (3)
  • Frame Interpolation (1)
  • Gal (1)
  • Gmail (1)
  • Google Forms (1)
  • Google-Apps-Script (1)
  • Google-Drive (1)
  • Gpu (1)
  • Graphs (1)
  • Group (1)
  • Group Management (1)
  • Group-Policy (1)
  • Gsuite (1)
  • Hacked Accounts (1)
  • Hardware (1)
  • Hex Encoding (1)
  • Iis (1)
  • Image-Processing (4)
  • Images (3)
  • Incident Response (1)
  • Insertion Sort (1)
  • Installation (1)
  • Interactive (9)
  • Ip-Address (1)
  • Ip-Addressing (1)
  • Java (1)
  • Javascript (5)
  • Juno (1)
  • Jupiter (1)
  • K-Means (1)
  • Kattis (6)
  • Keyboard (1)
  • Knowledge-Graphs (1)
  • Kruskal's Algorithm (1)
  • Lan (1)
  • Llm (2)
  • Local Administrator (1)
  • Local-Ai (2)
  • Logging (1)
  • Lorenz System (1)
  • M365 (3)
  • Machine-Learning (6)
  • Maximum Flow (1)
  • Media Processing (1)
  • Merge Sort (1)
  • Microsoft Teams (1)
  • Microsoft-Office (1)
  • Midas (1)
  • Minimum Spanning Tree (2)
  • Mistral-7b (1)
  • Monitoring (1)
  • Moondream (2)
  • Multilingual (1)
  • Mx (1)
  • N-Central (1)
  • Natural Language Processing (1)
  • Net (2)
  • Netsh (2)
  • Network (1)
  • Network Drives (1)
  • Network-Analysis (1)
  • Network-Security (2)
  • Networking (5)
  • Networkx (1)
  • Nlp (1)
  • Nslookup (1)
  • Obfuscation (1)
  • Office-365 (2)
  • Office365 (1)
  • Officec2rclient (1)
  • Open Simplex Noise (3)
  • Openai (1)
  • Optimization (1)
  • P5.js (2)
  • P5js (1)
  • Password Management (2)
  • Password-Generator (1)
  • Passwords (2)
  • Perlin Noise (1)
  • Permissions (2)
  • Phishing (1)
  • Photo-Editing (1)
  • Pil (1)
  • Pillow (1)
  • Port-Management (1)
  • Powershell (22)
  • Prim's Algorithm (1)
  • Prime Numbers (3)
  • Printers (1)
  • Procdump (1)
  • Processing (2)
  • Programming (2)
  • Python (15)
  • Python-Script (1)
  • Pyvis (2)
  • Qr-Code (1)
  • Rdp (1)
  • Reference (1)
  • Registry Modification (1)
  • Remote-Access (1)
  • Reporting (1)
  • Reports (3)
  • Robocopy (1)
  • Screen Recording (1)
  • Scripting (1)
  • SDK (1)
  • Security (9)
  • Security Analysis (1)
  • Security Management (1)
  • Settings (1)
  • Shell (4)
  • SID (1)
  • SMTP (2)
  • Sorting (3)
  • Sound (1)
  • Space (1)
  • Speech Recognition (1)
  • Spf (1)
  • Spiral (1)
  • Stable-Diffusion (2)
  • Stocks (1)
  • String (1)
  • Stub (1)
  • Subnets (1)
  • Synchronization (1)
  • Sysinternals (1)
  • System-Administration (13)
  • Systeminfo (1)
  • Team Management (1)
  • Team Ownership (1)
  • Tensorflow (1)
  • Therafit (1)
  • Time (1)
  • Topological Sort (1)
  • Troubleshooting (4)
  • Tzutil (1)
  • UDM (1)
  • Uri (1)
  • Uri Encoding (1)
  • User-Management (4)
  • Uva (9)
  • VBScript (1)
  • Version-Management (1)
  • Video (5)
  • Video Conversion (1)
  • Visualization (3)
  • Web-Administration (1)
  • Web-Development (1)
  • Wifi (1)
  • Win32_OperatingSystem (1)
  • Windbg (1)
  • Windows (17)
  • Windows 10 (1)
  • Windows 11 (1)
  • Windows-Defender (1)
  • Windows-Server (1)
  • Windows-Update (1)
  • Wmic (1)
  • Youtube-Dl (1)
  • Yt-Dlp (2)

© 2025 Ghostfeed theme by Tristan Madden. All rights reserved.