Omi SDKs
The Omi SDKs make it easy to build applications that connect directly to Omi devices. Connect via Bluetooth, stream audio, and get real-time transcription in your preferred language.
Available SDKs
Python SDK
pip-installable SDK with Opus decoding and Deepgram transcription
Swift SDK
Native iOS/macOS package with local Whisper transcription
React Native SDK
Cross-platform mobile SDK with BLE support
Key Features
Direct Device Connection
Connect to any Omi device over Bluetooth Low Energy
Flexible Transcription
Run transcription locally (Whisper) or via cloud (Deepgram)
No Backend Required
No dependency on Omi infrastructure - build completely standalone apps
Open Source
MIT licensed - use and modify freely
Quick Comparison
| Feature | Python | Swift | React Native |
|---|---|---|---|
| Installation | pip install omi-sdk | Swift Package Manager | npm install @omiai/omi-react-native |
| Transcription | Deepgram (cloud) | Whisper (local) | Deepgram (cloud) |
| Platforms | macOS, Linux, Windows | iOS, macOS | iOS, Android |
| Audio Codec | Opus | Opus | Opus, PCM |
Getting Started
1
Choose Your SDK
Pick the SDK that matches your development environment and use case.
2
Install Dependencies
Follow the installation guide for your chosen SDK.
3
Scan for Devices
Use the SDK’s scan function to find nearby Omi devices.
4
Connect & Stream
Connect to a device and start receiving audio or transcription data.