Azure Speech Services REST API v3.0 is now available
Published date: June 25, 2020
Azure Azure Speech Services REST API v3.0 is now available, along with several new features. Azure Speech Services is the unification of speech-to-text, text-to-speech, and speech-translation into a single Azure subscription. Easily enable any of the services for your applications, tools, and devices with the Speech SDK, Speech Devices SDK, or REST APIs.
New features include:
- Notifications-Webhooks—All running processes of the service now support webhook notifications. REST API v3.0 provides the calls to enable you to register your webhooks where notifications are sent.
- Update models behind endpoints—The API now updates the way model deployments are handled. Some scenarios require continued data gathering, periodic adaptation, and redeployment in regular intervals.
- Adapt with multiple data sets—Adapt a model using multiple data set combinations of acoustic, language, and pronunciation data with the updated API.
- Bring your own storage—Use your own storage accounts for logs, transcription files, and other data.
Use the updated REST API to:
- Copy models to other subscriptions in case you want colleagues to have access to a model you built, or in cases where you want to deploy a model to more than one region.
- Transcribe data from a container (bulk transcription) as well as provide multiple audio file URLs.
- Upload data from Azure Storage accounts through the use of a SAS Uri.
- Get logs per endpoint if logs have been requested for that endpoint.
- Request the manifest of the models you create, for the purpose of setting up on-premises containers.