Optional diagnostic data for Windows 11 and Windows 10

Applies to:

• Windows 11, version 23H2
• Windows 11, version 22H2
• Windows 11, version 21H2
• Windows 10, version 22H2
• Windows 10, version 21H2
• Windows 10, version 1809

Note

The information in this article also applies to these versions of Windows 10: 21H1, 20H2, 2004, 1909, 1903, 1803, and 1709. But those versions have reached their end of servicing date. For more information, see Microsoft Product Lifecycle.

Microsoft uses Windows diagnostic data to keep Windows secure and up-to-date, troubleshoot problems, and make product improvements. For users who have turned on “Tailored experiences”, it can also be used to offer you personalised tips, ads and recommendations to enhance Microsoft products and services for your needs. This article describes all types of optional diagnostic data collected by Windows, with comprehensive examples of data we collect per each type. For additional, detailed technical descriptions of Basic data items, see Windows 11 required diagnostic events and fields.

In addition, this article provides references to equivalent definitions for the data types and examples from ISO/IEC 19944-1:2020 Information technology - Cloud computing - Cloud services and devices: Data flow, data categories, and data use. Each data type also has a Data Use statement, for diagnostics and for Tailored experiences on the device, using the terms as defined by the standard. These Data Use statements define the purposes for which Microsoft processes each type of Windows diagnostic data, using a uniform set of definitions referenced at the end of this document and based on the ISO standard. Reference to the ISO standard provides additional clarity about the information collected, and allows easy comparison with other services or guidance that also references the standard.

The data covered in this article is grouped into the following types:

• Common data extensions (diagnostic header information)
• Device, Connectivity, and Configuration data
• Product and Service Usage data
• Product and Service Performance data
• Software Setup and Inventory data
• Browsing History data
• Inking, Typing, and Speech Utterance data

Common data extensions

Most diagnostic events contain a header of common data. In each example, the info in parentheses provides the equivalent definition for ISO/IEC 19944-1:2020.

Data Use for Common data extensions Header data supports the use of data associated with all diagnostic events. Therefore, Common data is used to provide Windows 10 and Windows 11, and may be used to improve, personalise, recommend, offer or promote Microsoft and third-party products and services, depending on the uses described in the Data Use statements for each data category.

Data Description for Common data extensions type

Common data extensions type

Information that is added to most diagnostic events, if relevant and available:

• Diagnostic level - Required or Optional, Sample level - for sampled data, what sample level is this device opted into (8.2.3.2.4 Observed Usage of the Service Capability)
• Operating system name, version, build, and locale (8.2.3.2.2 Telemetry data)
• Event collection time (8.2.3.2.2 Telemetry data)
• User ID - a unique identifier associated with the user's Microsoft Account (if one is used) or local account. The user's Microsoft Account identifier is not collected from devices configured to send Basic - diagnostic data (8.2.5 Account data)
• Xbox UserID (8.2.5 Account data)
• Device ID - This ID is not the user provided device name, but an ID that is unique for that device. (8.2.3.2.3 Connectivity data)
• Device class - Desktop, Server, or Mobile (8.2.3.2.3 Connectivity data)
• Environment from which the event was logged - Application ID of app or component that logged the event, Session GUID. Used to track events over a given period of time, such as the amount of time an app is running or between boots of the operating system (8.2.4 Cloud service provider data)
• Diagnostic event name, Event ID, ETW opcode, version, schema signature, keywords, and flags (8.2.4 Cloud service provider data)
• HTTP header information, including the IP address. This IP address is the source address that’s provided by the network packet header and received by the diagnostics ingestion service (8.2.4 Cloud service provider data)
• Various IDs that are used to correlate and sequence related events together (8.2.4 Cloud service provider data)

Device, Connectivity, and Configuration data

This type of data includes details about the device, its configuration and connectivity capabilities, and status. Device, Connectivity, and Configuration data is equivalent to ISO/IEC 19944-1:2020, 8.2.3.2.3 Connectivity data.

Data Use for Device, Connectivity, and Configuration data

For Diagnostics:
Pseudonymized Device, Connectivity, and Configuration data from Windows 10 and Windows 11 is used by Microsoft to provide and improve Windows 10 and Windows 11 and related Microsoft products and services. For example:

• Device, Connectivity, and Configuration data is used to understand the unique device characteristics that can contribute to an error experienced on the device, to identify patterns, and to more quickly resolve problems that impact devices with unique hardware, capabilities, or settings. For example:

  • Data about the use of mobile modems and their configuration on your devices is used to troubleshoot mobile modem issues.
  • Data about the use of USB hubs use and their configuration on your devices is used to troubleshoot USB hub issues.
  • Data about the use of connected Bluetooth devices is used to troubleshoot compatibility issues with Bluetooth devices.

• Data about device properties, such as the operating system version and available memory, is used to determine whether the device is due to, and able to, receive a Windows update.

• Data about device peripherals is used to determine whether a device has installed drivers that might be negatively impacted by a Windows update.

• Data about which devices, peripherals and settings are most-used by customers is used to prioritise Windows 10 and Windows 11 improvements to determine the greatest positive effect to the most Windows 10 and Windows 11 users.

With (optional) Tailored experiences:
If a user has enabled Tailored experiences on the device, Pseudonymised Device, Connectivity and Configuration data from Windows 10 and Windows 11 is used by Microsoft to personalise, recommend and offer Microsoft products and services to Windows 10 and Windows 11 users. Also, if a user has enabled Tailored experiences on the device, Pseudonymized Device, Connectivity, and Configuration data from Windows 10 and Windows 11 is used by Microsoft to promote third-party Windows apps, services, hardware, and peripherals to Windows 10 and Windows 11users. For example:

• Data about device properties and capabilities is used to provide tips about how to use or configure the device to get the best performance and user experience.

• Data about device capabilities, such as whether the device is pen-enabled, is used to recommend (Microsoft and third-party) apps that are appropriate for the device. These apps might be free or paid.
   

Data Description for Device, Connectivity, and Configuration data type

Device properties subtype: Information about the operating system and device hardware

• Operating system - version name, edition
• Installation type, subscription status, and genuine operating system status
• Processor architecture, speed, number of cores, manufacturer, and model
• OEM details - manufacturer, model, and serial number
• Device identifier and Xbox serial number
• Firmware/BIOS operating system - type, manufacturer, model, and version
• Memory - total memory, video memory, speed, and how much memory is available after the device has reserved memory
• Storage - total capacity and disk type
• Battery - charge capacity and InstantOn support
• Hardware chassis type, colour, and form factor
• Is this machine a virtual machine?

Device capabilities subtype: Information about the capabilities of the device

• Camera - whether the device has a front facing camera, a rear facing camera, or both.
• Touch screen - Does the device have a touch screen? If yes, how many hardware touch points are supported?
• Processor capabilities - CompareExchange128, LahfSahf, NX, PrefetchW, and SSE2
• Trusted Platform Module (TPM) - whether a TPM exists and if yes, what version
• Virtualisation hardware - whether an IOMMU exists, whether it includes SLAT support and whether virtualisation is enabled in the firmware
• Voice - whether voice interaction is supported and the number of active microphones
• Number of displays, resolutions, and DPI
• Wireless capabilities
• OEM or platform face detection
• OEM or platform video stabilisation and quality-level set
• Advanced Camera Capture mode (HDR versus Low Light), OEM versus platform implementation, HDR probability, and Low Light probability

Device preferences and settings subtype: Information about the device settings and user preferences

• User Settings - System, Device, Network & Internet, Personalisation, Cortana, Apps, Accounts, Time & Language, Gaming, Ease of Access, Privacy, Update & Security
• User-provided device name
• Whether device is domain-joined, or cloud-domain joined (for example, part of a company-managed network)
• Hashed representation of the domain name
• MDM (mobile device management) enrolment settings and status
• BitLocker, Secure Boot, encryption settings, and status
• Windows Update settings and status
• Developer Unlock settings and status
• Default app choices
• Default browser choice
• Default language settings for app, input, keyboard, speech, and display
• App store update settings
• Enterprise OrganizationID, Commercial ID

Device peripherals subtype: Information about the peripherals of the device

• Peripheral name, device model, class, manufacturer, and description
• Peripheral device state, install state, and checksum
• Driver name, package name, version, and manufacturer
• HWID - A hardware vendor-defined ID to match a device to a driver INF file
• Driver state, problem code, and checksum
• Whether driver is kernel mode, signed, and image size

Device network info subtype: Information about the device network configuration

• Network system capabilities
• Local or Internet connectivity status
• Proxy, gateway, DHCP, DNS details, and addresses
• Whether it's a paid or free network
• Whether the wireless driver is emulated
• Whether it's access point mode-capable
• Access point manufacturer, model, and MAC address
• WDI Version
• Name of networking driver service
• Wi-Fi Direct details
• Wi-Fi device hardware ID and manufacturer
• Wi-Fi scan attempt and item counts
• Whether MAC randomisation is supported and enabled
• Number of supported spatial streams and channel frequencies
• Whether Manual or Auto-connect is enabled
• Time and result of each connection attempt
• Aeroplane mode status and attempts
• Interface description provided by the manufacturer
• Data transfer rates
• Cypher algorithm
• Mobile Equipment ID (IMEI) and Mobile Country Code (MCCO)
• Mobile operator and service provider name
• Available SSIDs and BSSIDs
• IP Address type - IPv4 or IPv6
• Signal Quality percentage and changes
• Hotspot presence detection and success rate
• TCP connection performance
• Miracast device names
• Hashed IP address

Product and Service Usage data

This type of data includes details about the usage of the device, operating system, applications, and services. Product and Service Usage data is equivalent to ISO/IEC 19944-1:2020, 8.2.3.2.4 Observed Usage of the Service Capability.

Data Use for Product and Service Usage data

For Diagnostics:
Pseudonymized Product and Service Usage data from Windows 10 and Windows 11 is used by Microsoft to provide and improve Windows 10 and Windows 11 and related Microsoft product and services. For example:

• Data about the specific apps that are in-use when an error occurs is used to troubleshoot and repair issues with Windows features and Microsoft apps.
• Data about the specific apps that are most used by customers is used to prioritise Windows 10 and Windows 11 improvements to determine the greatest positive effect on the greatest number of Windows 10 and Windows 11 users.
• Data about whether devices have Suggestions turned off from the Settings Phone screen is to improve the Suggestions feature.
• Data about whether a user cancelled the authentication process in their browser is used to help troubleshoot issues with and improve the authentication process.
• Data about when and what feature invoked Cortana is used to prioritise efforts for improvement and innovation in Cortana.
• Data about when a context menu in the photo app is closed is used to troubleshoot and improve the photo app.

With (optional) Tailored experiences:
If a user has enabled Tailored experiences on the device, pseudonymised Product and Service Usage data from Windows 10 and Windows 11 is used by Microsoft to personalise, recommend and offer Microsoft products and services to Windows 10 and Windows 11 users. Also, if a user has enabled Tailored experiences on the device, pseudonymized Product and Service Usage data from Windows 10 and Windows 11 is used by Microsoft to promote third-party Windows apps, services, hardware, and peripherals to Windows 10 and Windows 11 users. For example:

• If data shows that a user has not used a particular feature of Windows, we might recommend that the user try that feature.
• Data about which apps are most-used on a device is used to provide recommendations for similar or complementary (Microsoft or third-party) apps. These apps might be free or paid.

Data Description for Product and Service Usage data type

App usage subtype: Information about Windows and application usage

• Operating system component and app feature usage
• User navigation and interaction with app and Windows features. This information could include user input, such as the name of a new alarm set, user menu choices or user favourites
• Time of and count of app and component launches, duration of use, session GUID, and process ID
• App time in various states –- running in the foreground or background, sleeping, or receiving active user interaction
• User interaction method and duration –- whether the user used a keyboard, mouse, pen, touch, speech, or game controller, and for how long
• Cortana launch entry point and reason
• Notification delivery requests and status
• Apps used to edit images and videos
• SMS, MMS, VCard, and broadcast message usage statistics on primary or secondary lines
• Incoming and outgoing calls and voicemail usage statistics on primary or secondary lines
• Emergency alerts are received or displayed statistics
• Content searches within an app
• Reading activity - bookmarked, printed, or had the layout changed

App or product state subtype: Information about Windows and application state

• Start Menu and Taskbar pins
• Online and offline status
• App launch state –- with deep-links, such as Groove launching with an audio track to play or MMS launching to share a picture
• Personalisation impressions delivered
• Whether the user clicked on, or hovered over, UI controls or hotspots
• User provided feedback, such as Like, Dislike or a rating
• Caret location or position within documents and media files - how much has been read in a book in a single session, or how much of a song has been listened to.

Purchasing subtype: Information about purchases made on the device

• Product ID, edition ID, and product URI
• Offer details - price
• Date and time an order was requested
• Microsoft Store client type - web or native client
• Purchase quantity and price
• Payment type - credit card type and PayPal

Log-in properties subtype: Information about logins on the device

• Log-in success or failure
• Log-in sessions and state

Product and Service Performance data

This type of data includes details about the health of the device, operating system, apps, and drivers. Product and Service Performance data is equivalent to ISO/IEC 19944-1:2020 8.2.3.2.2 EUII Telemetry data.

Data Use for Product and Service Performance data

For Diagnostics:
Pseudonymized Product and Service Performance data from Windows 10 and Windows 11 is used by Microsoft to provide and improve Windows 10 and Windows 11 and related Microsoft product and services. For example:

• Data about the reliability of content that appears in the Windows Spotlight (rotating lock screen images) is used for Windows Spotlight reliability investigations.
• Timing data about how quickly Cortana responds to voice commands is used to improve Cortana listening performance.
• Timing data about how quickly the facial recognition feature starts up and finishes is used to improve facial recognition performance.
• Data about when an application window fails to appear is used to investigate issues with application window reliability and performance.

With (optional) Tailored experiences:
If a user has enabled Tailored experiences on the device, pseudonymised Product and Service Performance data from Windows 10 and Windows 11 is used by Microsoft to personalise, recommend and offer Microsoft products and services to Windows 10 and Windows 11 users. Also, if a user has enabled Tailored experiences on the device, pseudonymized Product and Service Performance data from Windows 10 and Windows 11 is used by Microsoft to promote third-party Windows apps, services, hardware, and peripherals to Windows 10 and Windows 11 users.

• Data about battery performance on a device may be used to recommend settings changes that can improve battery performance.
• If data shows a device is running low on file storage, we may recommend Windows-compatible cloud storage solutions to free up space.
• If data shows the device is experiencing performance issues, we may provide recommendations for Windows apps that can help diagnose or resolve these issues. These apps might be free or paid.

Microsoft doesn’t use crash and hang dump data to personalise, recommend, offer or promote any product or service.

Data Description for Product and Service Performance data type

Device health and crash data subtype: Information about the device and software health

• Error codes and error messages, name and ID of the app, and process reporting the error
• DLL library predicted to be the source of the error - for example, xyz.dll
• System-generated files - app or product logs and trace files to help diagnose a crash or hang
• System settings, such as registry keys
• User-generated files - files that are indicated as a potential cause for a crash or hang. For example, .doc, .ppt, .csv files
• Details and counts of abnormal shutdowns, hangs, and crashes
• Crash failure data - operating system, operating system component, driver, device, and first-party and third-party app data
• Crash and hang dumps, including:
  • The recorded state of the working memory at the point of the crash
  • Memory in-use by the kernel at the point of the crash.
  • Memory in-use by the application at the point of the crash
  • All the physical memory used by Windows at the point of the crash
  • Class and function name within the module that failed.

Device performance and reliability data subtype: Information about the device and software performance

• User interface interaction durations - Start menu display times, browser tab switch times, app launch and switch times, and Cortana and Search performance and reliability
• Device on and off performance - device boot, shut-down, power on and off, lock and unlock times and user authentication times (fingerprint and face recognition durations)
• In-app responsiveness - time to set alarm, time to fully render in-app navigation menus, time to sync reading list, time to start GPS navigation, time to attach picture MMS, and time to complete a Microsoft Store transaction
• User input responsiveness - onscreen keyboard invocation times for different languages, time to show autocomplete words, pen or touch latencies, latency for handwriting recognition to words, Narrator screen reader responsiveness, and CPU score
• UI and media performance and glitches versus smoothness - video playback frame rate, audio glitches, animation glitches (stutter when bringing up Start), graphics score, time to first frame, play/pause/stop/seek responsiveness, time to render PDF, dynamic streaming of video from OneDrive performance
• Disk footprint - Free disk space, out of memory conditions, and disk score
• Excessive resource utilisation - components impacting performance or battery life through high CPU usage during different screen and power states
• Background task performance - download times, Windows Update scan duration, Microsoft Defender Antivirus scan times, disk defrag times, email fetch times, service start-up and state transition times, and time to index on-device files for search results
• Peripheral and devices - USB device connection times, time to connect to a wireless display, printing times, network availability and connection times (time to connect to Wi-Fi, time to get an IP address from DHCP etc.), smart card authentication times, automatic brightness, and environmental response times
• Device setup - first setup experience times (time to install updates, install apps, connect to network and so on), time to recognise connected devices (printer and monitor) and time to set up a Microsoft Account
• Power and Battery life - power draw by component (Process/CPU/GPU/Display), hours of time the screen is off, sleep state transition details, temperature and thermal throttling, battery drain in a power state (screen off or screen on), processes and components requesting power use while the screen is off, autobrightness details, time device is plugged into AC versus battery, and battery state transitions
• Service responsiveness - Service URI, operation, latency, service success and error codes, and protocol
• Diagnostic heartbeat - regular signal used to validate the health of the diagnostics system

Movies subtype: Information about film consumption functionality on the device

Note

This isn't intended to capture user viewing, listening, or habits.

• Video Width, height, colour palette, encoding (compression) type, and encryption type
• Instructions about how to stream content for the user - the smooth streaming manifest of content file chunks that must be pieced together to stream the content based on screen resolution and bandwidth
• URL for a specific two-second chunk of content if there is an error
• Full-screen viewing mode details

Music & TV subtype: Information about music and TV consumption on the device

Note

This isn't intended to capture user viewing, listening, or habits.

• Service URL for song being downloaded from the music service - collected when an error occurs to facilitate restoration of service
• Content type (video, audio, or surround audio)
• Local media library collection statistics - number of purchased tracks and number of playlists
• Region mismatch - User's operating system region and Xbox Live region

Reading subtype: Information about reading consumption functionality on the device

Note

This isn't intended to capture user viewing, listening, or habits.

• App accessing content and status and options used to open a Microsoft Store book
• Language of the book
• Time spent reading content
• Content type and size details

Photos app subtype: Information about photos usage on the device

Note

This isn't intended to capture user viewing, listening, or habits.

• File source data - local, SD card, network device, and OneDrive
• Image and video resolution, video length, file sizes types, and encoding
• Collection view or full screen viewer use and duration of view

On-device file query subtype: Information about local search activity on the device

• Type of query issued and index type (ConstraintIndex or SystemIndex)
• Number of items requested and retrieved
• File extension of search result with which the user interacted
• Launched item type, file extension, index of origin, and the App ID of the opening app
• Name of process calling the indexer and the amount of time to service the query
• A hash of the search scope (file, Outlook, OneNote, or IE history). The state of the indices (fully optimised, partially optimised, or being built)

Entitlements subtype: Information about entitlements on the device

• Service subscription status and errors
• DRM and licence rights details - groove subscription or operating system volume licence
• Entitlement ID, lease ID, and package ID of the install package
• Entitlement revocation
• Licence type (trial, offline versus online) and duration
• Licence usage session

Software Setup and Inventory data

This type of data includes software installation and update information on the device. Software Setup and Inventory Data is a subtype of ISO/IEC 19944-1:2020 8.2.3.2.4 Observed Usage of the Service Capability.

Data Use for Software Setup and Inventory data

For Diagnostics:
Pseudonymized Software Setup and Inventory data from Windows 10 and Windows 11 is used by Microsoft to provide and improve Windows 10 and Windows 11 and related Microsoft product and services. For example:

• Data about the specific drivers that are installed on a device is used to understand whether there are any hardware or driver compatibility issues that should block or delay a Windows update.
• Data about when a download starts and finishes on a device is used to understand and address download problems.
• Data about the specific Microsoft Store apps that are installed on a device is used to determine which app updates to provide to the device.
• Data about the antimalware installed on a device is used to understand malware transmissions vectors.

With (optional) Tailored experiences:
If a user has enabled Tailored experiences on the device, pseudonymised Software Setup and Inventory data from Windows 10 and Windows 11 is used by Microsoft to personalise, recommend and offer Microsoft products and services to Windows 10 and Windows 11 users. Also, if a user has enabled Tailored experiences on the device, pseudonymized Software Setup and Inventory data from Windows 10 and Windows 11 is used by Microsoft to promote third-party Windows apps, services, hardware, and peripherals to Windows 10 and Windows 11 users. For example:

• Data about the specific apps that are installed on a device is used to provide recommendations for similar or complementary apps in the Microsoft Store.

Data Description for Software Setup and Inventory data type

Installed applications and install history subtype: Information about apps, drivers, update packages, or operating system components installed on the device

• App, driver, update package, or component’s Name, ID, or Package Family Name
• Product, SKU, availability, catalogue, content, and Bundle IDs
• Operating system component, app or driver publisher, language, version and type (Win32 or UWP)
• Install date, method, install directory, and count of install attempts
• MSI package and product code
• Original operating system version at install time
• User, administrator, or mandatory installation or update
• Installation type - clean install, repair, restore, OEM, retail, upgrade, or update

Device update information subtype: Information about apps, drivers, update packages, or operating system components installed on the device

• Update Readiness analysis of device hardware, operating system components, apps, and drivers (progress, status, and results)
• Number of applicable updates, importance, and type
• Update download size and source - CDN or LAN peers
• Delay upgrade status and configuration
• Operating system uninstall and rollback status and count
• Windows Update server and service URL
• Windows Update machine ID
• Windows Insider build details

Browsing History data

This type of data includes details about web browsing in the Microsoft browsers. Browsing History data is equivalent to ISO/IEC 19944-1:2020 8.2.3.2.8 Client-side browsing history.

Data Use for Browsing History data

For Diagnostics:
Pseudonymized Browsing History data from Windows 10 and Windows 11 is used by Microsoft to provide and improve Windows 10 and Windows 11 and related Microsoft product and services. For example:

• Data about when the Block Content dialogue box has been shown is used for investigations of blocked content.
• Data about potentially abusive or malicious domains is used to make updates to Microsoft Edge and Windows Defender SmartScreen to warn users about the domain.
• Data about when the Address bar is used for navigation purposes is used to improve the Suggested Sites feature and to understand and address problems arising from navigation.
• Data about when a Web Notes session starts is used to measure popular domains and URLs for the Web Notes feature.
• Data about when a default Home page is changed by a user is used to measure which default Home pages are the most popular and how often users change the default Home page.

With (optional) Tailored experiences:
If a user has enabled Tailored experiences on the device, pseudonymised Browsing History data from Windows 10 and Windows 11 is used by Microsoft to personalise, recommend and offer Microsoft products and services to Windows 10 and Windows 11 users. Also, if a user has enabled Tailored experiences on the device, pseudonymized Browsing History data from Windows 10 and Windows 11 is used by Microsoft to promote third-party Windows apps, services, hardware, and peripherals to Windows 10 and Windows 11 users. For example:

• We might recommend that a user download a compatible app from the Microsoft Store if they have browsed to the related website. For example, if a user uses the Facebook website, we may recommend the Facebook app.

Data Description for Browsing History data type

Microsoft browser data subtype: Information about Address bar and Search box performance on the device

• Text typed in Address bar and Search box
• Text selected for an Ask Cortana search
• Service response time
• Autocompleted text, if there was an autocomplete
• Navigation suggestions provided based on local history and favourites
• Browser ID
• URLs (may include search terms)
• Page title

Inking Typing and Speech Utterance data

This type of data gathers details about the voice, inking, and typing input features on the device. Inking, Typing, and Speech Utterance data is a subtype of ISO/IEC 19944-1:2020 8.2.3.2.1 End User Identifiable information.

Data Use for Inking, Typing, and Speech Utterance data

For Diagnostics:
Anonymized Inking, Typing, and Speech Utterance data from Windows 10 and Windows 11 is used by Microsoft to improve natural language capabilities in Microsoft products and services. For example:

• Data about words marked as spelling mistakes and replaced with another word from the context menu is used to improve the spelling feature.
• Data about alternative words shown and selected by the user after right-clicking is used to improve the word recommendation feature.
• Data about autocorrected words that were restored back to the original word by the user is used to improve the autocorrect feature.
• Data about whether Narrator detected and recognised a touch gesture is used to improve touch gesture recognition.
• Data about handwriting samples sent from the Handwriting Panel is used to help Microsoft improve handwriting recognition.

With (optional) Tailored experiences:

Microsoft doesn't use Windows Inking, Typing, and Speech Utterance data for Tailored experiences.

Data Description for Inking, Typing, and Speech Utterance data type

Voice, inking, and typing subtype: Information about voice, inking, and typing features

• Type of pen used (highlighter, ball point, or pencil), pen colour, stroke height and width, and how long it is used
• Pen gestures (click, double click, pan, zoom, or rotate)
• Palm Touch x,y co-ordinates
• Input latency, missed pen signals, number of frames, strokes, first frame commit time, and sample rate
• Ink strokes written, text before and after the ink insertion point, recognised text entered, input language – processed to remove identifiers, sequencing information and other data (such as email addresses and numeric values), which could be used to reconstruct the original content or associate the input to the user
• Text to speech recognition results – result codes and recognised text
• Language and model of the recogniser and the System Speech language
• App ID using speech features
• Whether user is known to be a child
• Confidence and success or failure of speech recognition

ISO/IEC 19944-1:2020-specific terminology

This section provides the ISO/IEC 19944-1:2020-specific definitions for use and de-identification qualifiers used in this article.

Provide

ISO/IEC 19944-1:2020 Reference: 9.3.2 Provide

Use of a specified data category by a Microsoft product or service to protect and provide the described service, including, (i) troubleshoot and fix issues with the product or service or (ii) provide product or service updates.

Improve

ISO/IEC 19944-1:2020 Reference: 9.3.3 Improve

Use of a specified data category to improve or increase the quality of a Microsoft product or service. Those improvements may be available to end users.

Personalise

ISO/IEC 19944-1:2020 Reference: 9.3.4 Personalise

Use of the specified data categories to create a customised experience for the end user in any Microsoft product or service.

Recommend

ISO/IEC 19944-1:2020 Reference: 9.3.4 Personalise

“Recommend” means use of the specified data categories to Personalise (9.3.4) the end user’s experience by recommending Microsoft products or services that can be accessed without the need to make a purchase or pay money.

Use of the specified data categories give recommendations about Microsoft products or services the end user may act on where the recommendation is (i) contextually relevant to the product or service in which it appears, (ii) that can be accessed without the need to make a purchase or pay money, and (iii) Microsoft receives no compensation for the placement.

Offer

ISO/IEC 19944-1:2020 Reference: 9.3.5 Offer upgrades or upsell

Implies that the source of the data is Microsoft products and services, and the upgrades offered come from Microsoft products and services that are relevant to the context of the current capability. The target audience for the offer is Microsoft customers.

Specifically, use of the specified data categories to make an offer or upsell new capability or capacity of a Microsoft product or service that is (i) contextually relevant to the product or service in which it appears; (ii) likely to result in additional future revenue for Microsoft from end user; and (iii) Microsoft receives no consideration for placement.

Promote

ISO/IEC 19944-1:2020 Reference: 9.3.6 Market/advertise/promote

Use of the specified data categories to promote a product or service in or on a first-party Microsoft product or service.

Data identification qualifiers

Here are the data identification qualifiers and the ISO/IEC 19944-1:2020 reference:

• Pseudonymized Data 8.3.3 Pseudonymized data. Microsoft usage notes are as defined.
• Anonymized Data 8.3.5 Anonymized data. Microsoft usage notes are as defined.
• Aggregated Data 8.3.6 Aggregated data. Microsoft usage notes are as defined.