---
canonical: "https://www.softatlas.io/pyannoteai/products/precision-3-api"
name: "Precision-3 API"
provider: "pyannoteAI"
category: "Artificial Intelligence / Generative Media & Voice AI / Speech-to-Text APIs"
review_count: 0
ai_powered: true
website: "https://www.pyannote.ai/precision-3"
updated: "2026-09-29"
---

# Precision-3 API

By [pyannoteAI](/pyannoteai.md)

Precision-3 API offers a high-accuracy interface for speaker diarization and conversation intelligence, allowing integration with existing STT systems for enhanced audio processing.

## Features
- Identify and separate speakers in audio recordings
- Tune sensitivity parameters like 'vadSensitivity' and 'crosstalkSensitivity'
- Provide interpretable output with 'speakerProbability', 'speechProbability', and 'crosstalkProbability'
- Deploy with three speed-to-accuracy modes for on-premises use
- Reduce diarization error rate significantly compared to commercial alternatives
- Integrate with existing STT systems for enhanced audio processing

## Built for
DEVELOPERS, CUSTOMER_SUPPORT, OPERATIONS

Source: https://www.softatlas.io/pyannoteai/products/precision-3-api