Skip to main content
Remove background noise from human conversations in real-time

​
Overview

Real-world voice calls suffer from background noise that degrades audio quality and the user experience. Office chatter, traffic sounds, keyboard typing, and ambient noise make conversations difficult and unprofessional. The Noise Cancellation model removes unwanted background noise while preserving natural voice quality, delivering crystal-clear audio in any acoustic environment.

​
How It Works

1

Take audio input

Take the user’s speech as input from the audio pipeline.
2

Process through Sanas NC

Remove background noise while preserving the primary speaker’s voice.
3

Deliver clean audio

Send clean audio to the encoder/network for transmission.
4

Great call experience

Deliver a clear voice without distractions to the recipient.

​
Model Overview

​
Voice Isolation - General

General Purpose Noise Cancellation

Preserves: Primary speaker up to ~1m
Removes: Background voices & noise
Max Sample Rate: 24 kHz
Latency: ~40ms (streaming)
Use: General calls, voice agents, conferencing, gamingView technical details →
Key Features:
  • Real-time processing with minimal latency
  • Natural voice quality preservation
  • Works across various acoustic environments
  • Optimized for single-speaker scenarios
  • Low compute cost for high concurrency

​
Audio Sample

Example: The source audio sample was recorded in an outdoor setup with multiple speakers nearby and other ambient noise. Sanas Voice Isolation preserves the main speaker’s voice while eliminating all other voices and ambient noises. Source audio Enhanced audio (Voice Isolation)

​
Use Cases

Here are a few ideal use cases:
IndustryUse CaseBenefits
Contact CentersAgent-customer callsClear communication in noisy call centers
Remote WorkVideo conferencesProfessional audio from home offices
GamingVoice chatCrystal-clear team communication
TelemedicineDoctor-patient callsPrivate, distraction-free consultations
Voice AgentsAI assistantsAccurate speech recognition in any setting

​
Quick Start (30 seconds)

import sanas_remote_sdk

# Initialize SDK
sdk = sanas_remote_sdk.CreateRemoteSDK()
init_params = sanas_remote_sdk.InitParams()
init_params.remoteEndpoint = "your_endpoint"
init_params.accountId = "your_account_id"
init_params.accountSecret = "your_account_secret"
sdk.Initialize(init_params)

# Create Noise Cancellation processor
audio_params = sanas_remote_sdk.AudioParams()
audio_params.modelName = "VI_G_NC3.0"
audio_params.sampleRate = 16000  # Or up to 24kHz

processor, result = sdk.CreateAudioProcessor(audio_params, state_callback)

# Process audio in real-time
for chunk in audio_stream:
    clean_audio = processor.ProcessSamples(chunk)
    # Send clean_audio to encoder/network