Zebra Binding for Python

Zebra Translate

Made in Vancouver, Canada by Picovoice

Zebra is a lightweight, on-device neural machine translation engine. Zebra is:

• Private; All processing runs locally.
• Cross-Platform:
  • Linux (x86_64), macOS (x86_64, arm64), Windows (x86_64, arm64)
  • Raspberry Pi (3, 4, 5)

Compatibility

• Python 3.9+
• Runs on Linux (x86_64), macOS (x86_64, arm64), Windows (x86_64, arm64), and Raspberry Pi (3, 4, 5).

Installation

pip3 install pvzebra

AccessKey

Zebra requires a valid Picovoice AccessKey at initialization. AccessKey acts as your credentials when using Zebra SDKs. You can get your AccessKey for free. Make sure to keep your AccessKey secret. Signup or Login to Picovoice Console to get your AccessKey.

Usage

Create an instance of the engine and translate some text:

import pvzebra

handle = pvzebra.create(access_key='${ACCESS_KEY}', model_path='${MODEL_PATH}')

translation = handle.translate('${TEXT}')
print(translation)

Replace ${ACCESS_KEY} with yours obtained from Picovoice Console, ${MODEL_PATH} with a supported translation model located here and ${TEXT} with the text to translate. Finally, when done be sure to explicitly release the resources using handle.delete().

Translation Models

Zebra translation models are located here. The selected model decides the source and target translation languages.

The format of the model follows:

zebra_params_${SOURCE}_${TARGET}.pv

Where ${SOURCE} is the language code of the source language and ${TARGET} is the language code of the target language for the translation.

Demos

pvzebrademo provides command-line utilities for translating text using Zebra.