Home

Welcome to the Azure TTS (Text To Speech) Wiki

This wiki lists some tips of using Azure TTS service. Here you can find some common asked questions or issues raised, technical howtos and the release update notes for the new features. It is maintained by the Azure TTS engineering team and is meant to be supplement for the official Azure speech documentation.

Introduction

• Why Azure TTS?
• How Neural TTS works
• Where can I see the demo or app using Azure TTS service?
• Cloud vs Edge TTS
• Typical TTS Scenario
• Hybrid TTS Update History

Technical HowTos

• How to do integration with Azure TTS
• How to deploy TTS containers
• How to get request ID for debugging purpose
• How to get word time stamp events using Azure TTS
• How to use SSML
• How to use cognitive service TTS SR in IVR
• How to choose different audio output format
• How to call multiple voice in SSML
• How to scale TTS output duration to a given length
• How to synthesize a large file into audio files
• Using REST or SDK
• Optimize latency in server scenario
  • What is the latency of calling Azure TTS
  • How to reduce HTTP call latency to TTS service
  • The best practice to call TTS service in server scenario
  • How to increase TTS request limits?
  • How to do stress test
• How to onboard custom neural voices

Community Contributions

• (external) Call Azure TTS on command line
• (external) Call Azure TTS from PowerShell
• (external) R package mscstts as a client to call Azure TTS from R

How to get support

• open a github issue for programming issues
• post on stackoverflow
• open Azure ticket to investigate subscription specific issues.

Azure TTS Update Notes

• 2020.04 updates
• 2020.03 updates
• 2020.02 updates
• 2020.01 updates
• 2019.12 updates
• 2019.11 updates
• 2019.05 updates
• 2018.12 updates
• 2018.09 updates

If you have more questions or issues, please feel free to open issues. We will review and follow up.