Long Audio API shall be retired on 31 March 2027 – Migrate to the brand new batch synthesis API

On 31/03/2027 the Long Audio API of Azure AI Speech shall be retired. Please transition to the brand new batch synthesis API by that date.

We encourage you to make the change sooner to achieve the richer advantages of the brand new batch synthesis API and benefit from the higher worth. In addition to the acquainted options you’re utilizing, right here’s a fast comparability between the lengthy Audio API and substitute.

Long Audio API Features

Small Subset of Voices obtainable
Higher Latency
Limited Audio Output Formats supported
Does not assist superior options corresponding to phrase boundaries or background audios
More costly

Batch Synthesis API Features

All platform TTS voices obtainable
Lower latency
Same output codecs supported because the real-time synthesis API
Supports superior options corresponding to phrase boundaries and background audios throughout synthesis
Same worth because the real-time synthesis API

From now via 31/03/2027, you possibly can proceed to make use of the Long Audio API or transition to <the brand new batch synthesis API> earlier than 31 March 2027.

After 31/03/2027, the Long Audio API won’t be supported.

Required motion  

Follow the detailed steps to transition to the brand new batch synthesis API. View pricing right here.

Help and assist 

If you could have questions, get solutions from neighborhood specialists in Microsoft Q&A. If you could have a assist plan and also you want technical assist, create a assist request. <Please present particulars on tips on how to full the assist ticket.