azure_tts

TEN extension of azure Text to speech service, which converts input text into human like synthesized speech.

Features

• Text to speech
• Rich languages and voices support
• Optimized for Low latency
• Interruptible

API

property	type	default	description
azure_subscription_key	string	""	Azure Speech service subscription key
azure_subscription_region	string	""	Azure Speech service subscription region
azure_synthesis_voice_name	string	""	e.g., en-US-AdamMultilingualNeural, check more available voices in languages and voices support
prosody	string	""	Azure Speech prosody
language	string	""	Azure Speech language
role	string	""	Azure Speech role
style	string	""	Azure Speech style

Development

Build

Only linux/amd64 building inside container is supported at the moment.

# inside docker ghcr.io/ten-framework/ten_agent_build:0.3.2

task build

Refer to ci for more details.

Unit test

Misc