TEN extension of azure Text to speech service, which converts input text into human like synthesized speech.
- Text to speech
- Rich languages and voices support
- Optimized for Low latency
- Interruptible
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 |
Only linux/amd64
building inside container is supported at the moment.
# inside docker ghcr.io/ten-framework/ten_agent_build:0.3.3
task build
Refer to ci for more details.