Microsoft LUIS Benchmark

Microsoft Azure has two Cognitive Services which can be used to detect intent and entities from speech. Azure Language Understanding (LUIS) is a cloud service that enables building custom language models that can extract intent and entity information from textual conversational phrases. In addition, Azure Speech to Text can be used to transcribe audio files.

Prerequisites

• Ubuntu 20.04 (x86_64)
• Git
• Python
• PIP
• Azure Account

Usage

1. Clone the repository:

git clone https://github.com/Picovoice/speech-to-intent-benchmark.git

2. Install the dependencies:

pip3 install -r requirements.txt

3. Sign up for an Azure account.

4. In the Azure portal, create a new Language Understanding resource.

5. Sign up for a LUIS account.

6. Once you're in the LUIS console, import /data/luis/barista.json as your new app.

7. Train your LUIS app. Publish your app and choose Staging Slot as your publishing slot.

8. Create a Speech resource.

9. Create a custom speech model to improve Microsoft's speech-to-text accuracy with LUIS. In the Speech Studio, create a new project. Upload the provided corpus under /data/watson/corpus.txt as training data.

10. Deploy your speech model.

Result