--- title: Policy CSP - Speech description: Policy CSP - Speech ms.author: maricia ms.topic: article ms.prod: w10 ms.technology: windows author: nickbrower ms.date: 07/14/2017 --- # Policy CSP - Speech > [!WARNING] > Some information relates to prereleased product which may be substantially modified before it's commercially released. Microsoft makes no warranties, express or implied, with respect to the information provided here.
## Speech policies **Speech/AllowSpeechModelUpdate**
Home Pro Business Enterprise Education Mobile Mobile Enterprise
check mark1 check mark1 check mark1 check mark1 check mark1 check mark1

Added in Windows 10, version 1607. Specifies whether the device will receive updates to the speech recognition and speech synthesis models. A speech model contains data used by the speech engine to convert audio to text (or vice-versa). The models are periodically updated to improve accuracy and performance. Models are non-executable data files. If enabled, the device will periodically check for updated speech models and then download them from a Microsoft service using the Background Internet Transfer Service (BITS).

The following list shows the supported values: - 0 – Not allowed. - 1 (default) – Allowed.


Footnote: - 1 - Added in Windows 10, version 1607. - 2 - Added in Windows 10, version 1703. - 3 - Added in Windows 10, version 1709.