Azure Text-to-Speech with Avatar

This Streamlit app allows you to submit and monitor batch synthesis jobs using Azure Text-to-Speech with Avatar. It leverages the Azure AI Services to create talking avatars based on the provided text.

Usage

• Choose the language and avatar style using the sidebar dropdowns.
• Type the text you want to synthesize in the selected language.
• Click the "Submit Synthesis Job" button to initiate the batch synthesis job.
• Monitor the status in real-time. Once the job is successful, the talking avatar video will be displayed.

Supported Languages and Avatars

The application supports multiple languages, each associated with a specific talking avatar style.

Language	Voice
Arabic	ZariyahNeural
Bahasa Indonesian	GadisNeural
Bengali	TanishaaNeural
Chinese Mandarin	XiaoxiaoNeural
Dutch	FennaNeural
English	AvaNeural
French	DeniseNeural
German	KatjaNeural
Hindi	SwaraNeural
Italian	ElsaNeural
Japanese	NanamiNeural
Korean	SunHiNeural
Russian	SvetlanaNeural
Spanish	ElviraNeural
Telugu	ShrutiNeural

Demo

Check out the demo video:

Demo.Video.mp4

Setup

To run the application, you need to set up your Azure Text-to-Speech subscription key, service region, and service host. You can set these values in a .env file or directly in the script.

The text-to-speech avatar feature is only available in these service regions: West US 2, West Europe, and Southeast Asia. Create a resource group in these regions and use the same.

SUBSCRIPTION_KEY=<your_subscription_key>
SERVICE_REGION=<your_service_region>

References

Learn more about Text-to-Speech Avatar on Microsoft Azure here