Skip to main content

Documentation Index

Fetch the complete documentation index at: https://developer.sanas.ai/llms.txt

Use this file to discover all available pages before exploring further.

Sanas delivers world-class AI models across two categories: Human ↔ Human and Human ↔ Machine. Sanas currently offers Noise Cancellation and Speech Enhancement capabilities. Accent Translation, Language Translation, and Speech Intelligence coming soon.

​
Noise Cancellation Models

ModelUse CaseDescriptionBest For
VI_G_NC3.0
NC — Voice Isolation (General)		Human ↔ HumanIsolates intended speech by removing background noise and voicesContact centers, conferencing, gaming
AGENTIC_VI_G_NC
Agentic NC — Voice Isolation (General)		Human ↔ MachineRemoves all background noise and non-primary voices for complete speaker isolationSingle-speaker isolation for voice agents, IVR, phone bots
AGENTIC_VI_GT_NC
Agentic NC — Voice Isolation (Telephony)		Human ↔ MachineTelephony-optimized variant of VI_G for 8kHz narrowband audioTelephony voice agents, IVR, contact centers
AGENTIC_ST_NC
Agentic NC — Standard		Human ↔ MachineRemoves background noise while keeping all human speech audibleMulti-speaker environments where background conversations carry context

​
Speech Enhancement Models

ModelUse CaseDescriptionBest For
SE2.1
SE Standard		Human ↔ HumanRestores and enhances voice quality for telephony audio, outputting 8kHzContact centers, telephony, IVR systems
SE2.2
SE Ultra		Human ↔ HumanFull-fidelity speech enhancement with bandwidth extension to 24kHzPremium contact centers, conferencing, telemedicine
Hear samples and learn more about each model’s specifications, use cases, and code examples:

​
Human ↔ Human

Noise Cancellation · Voice Isolation (General)

VI_G_NC3.0 — Isolates intended speech by removing background noise and voices. Optimized for human listeners.
  • Latency: ~40ms
  • Sample rate: Up to 24kHz
  • Range: Primary speaker within ~1m

Speech Enhancement · Standard

SE2.1 — Restores and enhances voice quality for telephony audio. Low CPU footprint.
  • Latency: 120ms
  • Sample rate: 16kHz → 8kHz

Speech Enhancement · Ultra

SE2.2 — Full-fidelity speech enhancement with bandwidth extension to ultra-fidelity 24kHz.
  • Latency: 160ms
  • Sample rate: 16kHz → 24kHz

​
Human ↔ Machine

Agentic Noise Cancellation · Voice Isolation (General)

AGENTIC_VI_G_NC — Removes background noise and distant voices for complete voice isolation of the primary speaker’s audio stream.
  • Latency: ~100ms
  • Sample rate: 16kHz
  • Relative Word Error Rate Reduction (RWERR): 5-30% (average)

Agentic Noise Cancellation · Voice Isolation (Telephony)

AGENTIC_VI_GT_NC — Telephony-optimized variant of Voice Isolation for 8kHz narrowband audio.
  • Latency: ~100ms
  • Sample rate: 8kHz
  • Relative Word Error Rate Reduction (RWERR): 5-30% (average)

Agentic Noise Cancellation · Standard

AGENTIC_ST_NC — Removes background noise while preserving all human speech for multi-speaker environments.
  • Latency: ~100ms
  • Sample rate: 16kHz
  • Relative Word Error Rate Reduction (RWERR): 5-30% (average)