Microsoft Speech Service API: Text-to-Speech Samples
Microsoft Text to speech service now is offically supported by Speech SDK
now. It is recommended way to use TTS in your service or apps. It
supports both REST and Websocket connection to the service. This repo
still contains sample in various languages for using Text-to-Speech. We
will also support the common development questions using the issues
tracker.
Speech Service is generally avaliable since 2018.9. Please be sure you have a subscription key for the new Speech Service, not a subcription key from the retiring Bing Speech.
Microsoft also offers Neural TTS GA which can be invoked following the samples in this repo as well. What you need is to use a neural TTS endpoint.
Neural TTS uses Deep Neural Networks in matching patterns of stress
and intonation in spoken language called prosody. So, it does prosody
prediction and voice synthesis simultaneously. While traditional TTS
break down prosody into separate linguistic analysis and the predictions
are governed by independent models, that results in obscure voice
synthesis. And therefore, Neural TTS perform much better than
traditional TTS.
Recommend to run the CSharp example first which is always kept up to date.
Contributing
We welcome contributions. Feel free to file issues and pull requests
on the repo and we'll address them as we can. Learn more about how you
can help on our Contribution Rules & Guidelines.
You can reach out to us anytime with questions and suggestions using our communities below:
All Microsoft Cognitive Services SDKs and samples are licensed with the MIT License. For more details, seeLICENSE.
Sample images are licensed separately, please refer to LICENSE-IMAGE.
Developer Code of Conduct
Developers using Cognitive Services, including this client library
& sample, are expected to follow the “Developer Code of Conduct for
Microsoft Cognitive Services”, found at http://go.microsoft.com/fwlink/?LinkId=698895.