run NLC and NLU algorithms on the device, we can provide similar

functionality without relying on cloud inference. I thought it would

be helpful to implement an Open Source project to implement basic

intent understanding functionalities, directly importing datasets of

intents and sample utterances designed with those popular NLU cloud

platforms.

SwiftNLC

SwiftNLC is a Natural Language Classier based on Core ML /

TensorFlow integration, capable of running oine on

iOS/watchOS/tvOS devices.

This project is available on GitHub at:

SwiftNLC is composed of dierent sub-projects:

Importer: A Swift multiplatform console app to import intents

and utterances from dierent formats (could run on macOS and

Linux)

SampleDatasets: A folder for JSON les containing intents

denitions and sample utterances

Embedder: A Swift macOS console app to prepare the word

embedding encoding using NSLinguisticTagger. It must run on

macOS and not on Linux, as Apple NSLinguisticTagger is not part

of the public multi-platform Foundation Library

ModelNotebook: A folder containing Jupyter Notebooks for

implementing the Deep Neural Network Classier model using

Keras/TensorFlow API and exporting it using the Apple

CoreMLTools python library

Wrapper: A Swift wrapper to the auto-generated Core ML model

to simplify access to the Core ML Classier model and

prepare/encode the utterances to use for prediction

JacopoMangiavacchi/SwiftNLC

SwiftNLC — Swift Natural Language Classier with

CoreML / TensorFlow

github.com

•

•

•

•

•