We are excited to announce a pricing adjustment to our speech to text offerings for our Azure Speech Service.
Standard Batch Pricing Update
Starting today the pricing for our Standard Batch services has been revised from $1.00/hr to $0.36/hr. This change allows you to enjoy the same high-quality service at a fraction of the previous cost. This new pricing applies when you use the new Speech to text REST API v3.2 preview we released today.
Custom Speech Batch Pricing Update
On October 1st, 2023, we will also update our Custom Speech Batch pricing from $1.40/hr to $0.45/hr. Here too, the new pricing applies when you use the new Speech to text REST API v3.2 preview.
Enhanced Add-Ons Included in Azure Speech Batch
We understand the importance of providing you with comprehensive solutions. As a token of appreciation for your continued support, we are now including the Enhanced Add-Ons (Language Identification and Diarization) as part of both the Standard Batch and Custom Batch offerings, at no extra charge. This change is independent of the API version used.
Real-time Custom Speech Pricing Update
On October 1st, 2023, we are lowering the pricing of our real-time custom speech to text service from $1.40/hr to $1.20/hr. This change will apply to all custom speech usage of our real-time speech to text service independent of the SDK version.
Custom Speech Model Training Pricing Update
When adapting with base models released on October 1st, 2023, or later we will start charging for Custom Speech model training which allows you to adapt our speech to text models to your scenario/domain. Custom Speech model training will be charged at $10.00 per compute hour. Both the Custom Speech portal in Speech Studio and the new Rest API v3.2 preview will provide information on which models will incur training costs.
Using the REST API v3.2 preview
In order to use the new API, you must update the base path in your code from /speechtotext/v3.1 to /speechtotext/v3.2-preview.1. For example, to get base models in the eastus region, use https://eastus.api.cognitive.microsoft.com/speechtotext/v3.2-preview.1/models/base instead of https://eastus.api.cognitive.microsoft.com/speechtotext/v3.1/models/base.
For more information on how to migrate to the new v3.2 preview API see Migrate from v3.1 to v3.2 REST API - Speech service - Azure AI services | Microsoft Learn.
If you'd like to learn more about how these updates can benefit your projects, please visit our pricing page to get more details about the coming price change.
