Azure Cognitive Services

Customer Feedback & Ideas for Azure Cognitive Services

Share your ideas for making Cognitive Services and the accompanying APIs work better for the applications you develop.


Catch up on the latest News and Updates


Share your Ideas and Feedback

To share your ideas on how we can make Cognitive Services better, click one of the categories underneath "Give Feedback" located in the sidebar menu to access the forum.


Documentation

API documentation available here. Within, you'll find:

§  Getting started samples
§  API References
§  Testing Consoles

Using one or more of the APIs as a "Free" preview?  Be sure to read our Terms of Service.

Contact Support

UserVoice is intended for product feedback. If you need product support, please contact either: Azure support (https://azure.microsoft.com/en-us/support/plans/) or ask a question on stack overflow (https://stackoverflow.com/questions/tagged/microsoft-cognitive)


Become a Cloud Design Insider!

Join Cloud Design Insiders, and help shape the future of Cognitive Services! As an insider, you’ll speak with program managers, designers & researchers, see new designs and ideas, provide feedback through surveys, and try out prototypes. Take the short survey to join the Cloud Design Insiders now, and we’ll see you in the community.


  1. After tagging an image the UI should return to the same page you left from, rather than resetting to Page 1.

    e.g. if you tag an image on page 630 of your selected Training Images as soon as you close the tagging dialog box the page view resets to Page 1. To get back to where you were tagging now requires hundreds of clicks to get back to the same page you were tagging, to continue.

    After the tagging dialog box the UI should return to the training images page you left.

    7 votes

    We're glad you're here

    Please sign in to leave feedback

    Signed in as (Sign out)

    We’ll send you updates on this idea

    1 comment  ·  Vision  ·  Flag idea as inappropriate…  ·  Admin →
  2. "Visualization of the current progress of the training" and "Function to stop training" in Custom Vision

    If the following features are available, training can be executed with Cutom Vision without any worries.


    • Visualization of the current progress of the training

    • Function to stop training

    6 votes

    We're glad you're here

    Please sign in to leave feedback

    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Vision  ·  Flag idea as inappropriate…  ·  Admin →
  3. Generate accurate audio clip for each utterance

    Getting an audio clip for each utterance will make it possible to generate a basis for a human-labeled transcript for training a custom model. This will make it possible to gradually improve the recognition accuracy after every "session", by checking the transcription and the corresponding audio clip and fixing the text for incorrect transcriptions.

    Additionally the audio clip can be used as a live read-back of the original audio.

    6 votes

    We're glad you're here

    Please sign in to leave feedback

    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Speech  ·  Flag idea as inappropriate…  ·  Admin →
  4. Add «cheque» to prebuilt models

    It would be really useful to have a prebuilt model for cheques, which are much more bound to a standard than current prebuilt models (invoices, business cards and receipts).

    6 votes

    We're glad you're here

    Please sign in to leave feedback

    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Vision  ·  Flag idea as inappropriate…  ·  Admin →
  5. Magnetic Ink Character Recognition (MICR) for cheques

    Right now, Azure Form Recognizer mistakenly interpret MICR special encoding characters as regular characters, as you can see in the attached screenshot. It would be very userful that it ignore/handle those special characters.

    6 votes

    We're glad you're here

    Please sign in to leave feedback

    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Vision  ·  Flag idea as inappropriate…  ·  Admin →
  6. Need the new metric to check the number of characters used for text to speech on the Azure portal

    It is needed to be able to check the number of characters used for text to speech.
    Under the metrics tab on the Azure portal, we can only see the number of requests that have been made.

    4 votes

    We're glad you're here

    Please sign in to leave feedback

    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Speech  ·  Flag idea as inappropriate…  ·  Admin →
  7. Add Go to End '>>', and Go to the Beginning '<<' action for Training Images displayed in Custom Vision.

    When there are many thousands of images tagged in the Training Images view of a Project there is no quick way to get the the end of the data set, or back to the beginning. It can take a great many clicks, 60 images at a time.

    A Go to End '>>', and Go to the Beginning '<<' action on this interface would help.

    4 votes

    We're glad you're here

    Please sign in to leave feedback

    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Vision  ·  Flag idea as inappropriate…  ·  Admin →
  8. Speaker diarization for more than 2 speakers

    Speaker diarization for more than 2 speakers.

    See this one: https://cognitive.uservoice.com/forums/555925-speaker-recognition/suggestions/34823824-add-support-for-speaker-diarization-for-untrained

    I dont feel this should be marked as resolved. Would expect support for at least 10 speakers. Additionally its currently really poor and switches between speaker 1 and 2 almost randomly. Please make this more intelligent. Its a deal breaker for us and I'm sure many others. Especially considering the google alternative can handle unlimited speakers and is far more accurate at identifying them.

    https://cloud.google.com/speech-to-text/docs/multiple-voices

    And no... expecting a sample to train it for each voice is not an option. We literally just need it to assign a number…

    4 votes

    We're glad you're here

    Please sign in to leave feedback

    Signed in as (Sign out)

    We’ll send you updates on this idea

    1 comment  ·  Speech  ·  Flag idea as inappropriate…  ·  Admin →
  9. Form Recognizer - Mark all unused tags "unavailable in document"

    Form recognizer tagging Orders to the Common Data Model leaves me with many fields that are blank in certain Collections.
    Manually unchecking each filed "Unavailable in document" really slows the mapping down.

    Consider adding Multiselect "Unavailable in document" during tagging or better yet, Mark all unused tags "unavailable in document"

    3 votes

    We're glad you're here

    Please sign in to leave feedback

    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Decision  ·  Flag idea as inappropriate…  ·  Admin →
  10. Audio Offset / Duration for Best Result on normalized words

    The JSON and/or result object needs to have the offset and duration of the whole normalized word.
    I've reviewed the JSON and it still doesn't solve the problem. I need to know the relationship of the DisplayText words to the Word Timings in the detail When the DisplayText outputs 007 and the Word Timings output "double" "oh" "seven" as 3 different words I don't know that 007 = those three words as there is no reference. There needs to be a display word reference to the audio word to track offset/duration of an underlying audio file. The only option that…

    2 votes

    We're glad you're here

    Please sign in to leave feedback

    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Speech  ·  Flag idea as inappropriate…  ·  Admin →
  11. Create model that recognises a frame or video clip using a model trained with videos

    Azure computer vision provides an impressive and helpful image recognition service at the following web site;-
    https://portal.azure.com/#create/Microsoft.CognitiveServicesCustomVision
    It would be very valuable to users if Azure was to provide a service that can detect and classify an uploaded image by comparing that with uploaded video clips. Basically have a service that trains a model using video clips, not individual images, and then predicts when presented with another image or video clip, using the same simple and helpful interface that already exists.

    1 vote

    We're glad you're here

    Please sign in to leave feedback

    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Vision  ·  Flag idea as inappropriate…  ·  Admin →
  12. Allow System Managed Identity to access private Blob Storage

    I would like to be able to read files in private Blob storage with the managed identity of my Computer Vision Cognitive Service. The account is granted Reader and Storage Blob Data Reader, but still has no Access to the Blob unless I make it fully public.

    1 vote

    We're glad you're here

    Please sign in to leave feedback

    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Vision  ·  Flag idea as inappropriate…  ·  Admin →
  13. Add support to turn answers on and off

    For content that changes regularly, where it could be a "yes" one day and "no" the next. Instead of us having to either re-write these QnA pairs or edit them when they are potentially going to flip back, is it possible to ‘switch off’ certain answers so they still are on the KB but not visible when we publish them?

    1 vote

    We're glad you're here

    Please sign in to leave feedback

    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Decision  ·  Flag idea as inappropriate…  ·  Admin →
  14. Read API recognising some text twice

    Fix a bug within the Azure Computer Vision Read API 3.0 that leads to some text being recognised twice.

    An example is attached.

    1 vote

    We're glad you're here

    Please sign in to leave feedback

    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Vision  ·  Flag idea as inappropriate…  ·  Admin →
  15. Timeline tagging functionality

    Is there a way to add tag to a specific time code in the video?
    We already have the transcript generated but if we wanted to go to a specific time in the video, is that possible?

    1 vote

    We're glad you're here

    Please sign in to leave feedback

    Signed in as (Sign out)

    We’ll send you updates on this idea

    3 comments  ·  Video Indexer  ·  Flag idea as inappropriate…  ·  Admin →
  16. Content Moderator needs a scale out capability

    Max scale for content moderator looks to only be 10/sec. It would be better to have a scale out capability that allows for scaling up instances per load and scaling down when the load goes down.

    1 vote

    We're glad you're here

    Please sign in to leave feedback

    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  General Cognitive Services Feedback  ·  Flag idea as inappropriate…  ·  Admin →
  17. Allow alternative phrasing in docx source for QNA Maker

    You can do alternative phrasing in an Excel file by simply repeating the answer multiple times with different questions. It would be nice to be able to do the same by simply typing multiple questions with H1 formatting before typing the answer in a docx file. End users are much more comfortable editing word documents than Excel files.

    Example:

    What is our street address?
    What is our address?
    What is the company's physical address?
    One Microsoft Way
    Redmond, WA 98052

    1 vote

    We're glad you're here

    Please sign in to leave feedback

    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  General Cognitive Services Feedback  ·  Flag idea as inappropriate…  ·  Admin →
  18. barcode recognition

    Please extend the Read API so that it can read 2d-barcodes.

    1 vote

    We're glad you're here

    Please sign in to leave feedback

    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Vision  ·  Flag idea as inappropriate…  ·  Admin →
  19. FormRecogniser setup institutions incomplete

    I am implementing the Formrecogniser code in Python using the Microsoft quickstart guides.The code is unable to load azure API and gives a ModulenotFound error.
    I have the Azure SDK installed but my python code cannot find the path.
    Any pointers how to resolve the path issue.

    from azure.core.exceptions import ResourceNotFoundError
    ModuleNotFoundError: No module named 'azure'

    https://docs.microsoft.com/en-us/azure/cognitive-services/form-recognizer/quickstarts/client-library?tabs=ga%2Cv2-0&pivots=programming-language-python

    1 vote

    We're glad you're here

    Please sign in to leave feedback

    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Decision  ·  Flag idea as inappropriate…  ·  Admin →
  20. Site banner when there is a known issue

    Twice now the Speech portal has been broken by the owning Product Group.

    Twice now I have wasted hours of my time as well as MS support personnel time trying to debug something only to find out that the portal (and associated APIs) were broken and it was known by the group.

    Twice now the fix has been weeks in the deploying so god knows how many other customer's time has been wasted.

    If you have a known issue that affects your customers, especially given the woeful error messaging on the portal, then please add a banner on the…

    1 vote

    We're glad you're here

    Please sign in to leave feedback

    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Speech  ·  Flag idea as inappropriate…  ·  Admin →
← Previous 1 3
  • Don't see your idea?

Feedback and Knowledge Base