Microsoft

General Feedback

Thanks for using Cognitive Services! Which APIs are your favorite? How can we improve Cognitive Services?


                               Attention!





We have moved our Customer Feedback & Ideas for Azure Cognitive Services portal to the Azure Feedback Forum.





Please go to the link below to access our new Feedback and Ideas Page.



  • Hot ideas
  • Top ideas
  • New ideas
  • My feedback
  1. Support testing dispatched apps on luis.ai

    When combining Luis apps to a dispatched app it's not easy to test specific questions. So please support the same test functions for Dipatched apps like for "pure" Luis apps. Thank you.

    1 vote
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Flag idea as inappropriate…  ·  Admin →
  2. Support different font sizes in the same line (for ex. smaller digit after the decimal point)

    A lot of LCD displays smaller digits after the decimal point. Currently your algorithm fails to detect the text. I believe that it is quite easy to solve because all characters including the decimal point have the same "bottom" height.
    I'm attaching ex. photo

    1 vote
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Sample Requests  ·  Flag idea as inappropriate…  ·  Admin →
  3. Why is the API documentation specific to regions?

    The API documentation URLs are region-specific. For example:

    https://eastus.dev.cognitive.microsoft.com/docs/services/TextAnalytics.V2.0/operations/56f30ceeeda5650db055a3c7

    OR

    https://westcentralus.dev.cognitive.microsoft.com/docs/services/computer-vision-v3-ga/operations/5d986960601faab4bf452005

    I have been able to find one root URL for all APIs:

    https://dev.cognitive.microsoft.com/docs/services/

    But the above URL does not appear in any search results, either in Google or even in Bing. Also, there are no links to it from any of the landing pages for Cognitive Services.

    An API directory for Cognitive Services that is independent of regions would be welcome.

    1 vote
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Flag idea as inappropriate…  ·  Admin →
  4. False positives on Vision api isAdultContent and isRacyContent

    The attached image of a woman was rated as adult and racy by the Vision v3 api, and it really isn't.

    See e.g. request-id: 17f479bb-25a9-4acf-919a-ac91955d4fb3

    1 vote
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Other  ·  Flag idea as inappropriate…  ·  Admin →
  5. 1 vote
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Other  ·  Flag idea as inappropriate…  ·  Admin →
  6. Export to Excel should work

    Get Export and import to work in QnA maker - does not work as many writes on the net!

    1 vote
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Other  ·  Flag idea as inappropriate…  ·  Admin →
  7. Computer vision read api 3 public preview complains about image being too lart

    Hi, according to the public review's documentation an image size needs to be between 50 x 50 and 10000 x 10000 pixels. PDF pages must be 17 x 17 inches or smaller and the file size of the image must be less than 20 megabytes (MB).

    https://docs.microsoft.com/en-us/azure/cognitive-services/computer-vision/concept-recognizing-text#read-api

    I have a file that satisfies all requirements yet still faces this error:

    "error": {

        "code": "InvalidImageSize",
    
    "message": "The input image is too large. It should not be larger than 20MB."
    }

    I try creating another file with the same resolution but with fewer texts. This time it works. Could you please have…

    1 vote
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  UserVoice  ·  Flag idea as inappropriate…  ·  Admin →
  8. 1 vote
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Keys  ·  Flag idea as inappropriate…  ·  Admin →
  9. Translator Text API 3.0 dictionary lookup should support posTag as search criteria

    I'm using Microsoft's Translator Text API 3.0 dictionary lookup.
    The lookup of certain words is not correct because there seems to be no way to use posTag (ADJ, VERB, NOUN, ...) as search criteria.
    Here is an example:

    'https://api.cognitive.microsofttranslator.com/dictionary/lookup?api-version=3.0&from=de&to=en'
    with reqest Body: "[{"Text":"Arm","posTag":"NOUN"}]"

    This lookup retrieves only adjectives, and returns "poor" instead of "arm".
    posTag will not be taken into account.

    {{
    "normalizedSource": "arm",
    "displaySource": "arm",
    "translations": [

    {
    
    "normalizedTarget": "poor",
    "displayTarget": "poor",
    "posTag": "ADJ",
    "confidence": 1.0,
    "prefixWord": "",
    "backTranslations": [
    {
    "normalizedText": "arm",
    "displayText": "arm",
    "numExamples": 1,
    "frequencyCount": 9769
    }
    ]
    }

    ]
    }}

    1 vote
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  UserVoice  ·  Flag idea as inappropriate…  ·  Admin →
  10. 1 vote
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Flag idea as inappropriate…  ·  Admin →
  11. Include List entities in batch testing

    I am trying to run batch tests for my LUIS app. The entities labelled as List Entity are displayed as False Positive inside the chart. I have to exclude all the utterances containing the list entity while creating the batch file.
    It would be great if I would be able to test the Intent prediction for those utterances using the batch tests.

    I have posted in stack overflow regarding the same.
    https://stackoverflow.com/questions/60221072/batch-testing-in-luis-incorrect-prediction-for-list-enities/60230988#60230988

    9 votes
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Other  ·  Flag idea as inappropriate…  ·  Admin →
  12. Azure Cognitive Search

    Hi all, I would like to suggest to have sample source code for integrate Azure Cognitive Search with Azure Bot Services.

    1 vote
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Sample Requests  ·  Flag idea as inappropriate…  ·  Admin →
  13. Offline cognitive containers

    https://docs.microsoft.com/en-us/azure/cognitive-services/cognitive-services-container-support

    This article mentions "The Cognitive Services containers are required to submit metering information for billing purposes. The only exception, is Offline containers as they follow a different billing methodology." What exactly is the Offline containers.

    3 votes
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  UserVoice  ·  Flag idea as inappropriate…  ·  Admin →
  14. Customer suggests to have the per-device way of change of LUIS. As customer has around 3000 chat robot, each robot will talk a lot everyday.

    Customer suggests to have the per-device way of change of LUIS. As customer has around 3000 chat robot, each robot will talk a lot everyday.

    1 vote
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  UserVoice  ·  Flag idea as inappropriate…  ·  Admin →
  15. Add Pashto to Cognitive Services!

    This idea was submitted in Bing Translator with about 24,000 votes now, as the top and most requested. https://microsofttranslator.uservoice.com/forums/421776-how-do-i-get-my-language-supported/suggestions/14808924-pashto-afghanistan The inclusion of Pashto is significantly important for all international organizations working in Afghanistan and Pakistan.

    1 vote
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Language Support  ·  Flag idea as inappropriate…  ·  Admin →
  16. Translating text in Microsoft translator removes //

    0

    I am trying to translate a code file and I see the below issues with Microsoft translator

    Translating below text // 行番号削除正常終 Gives output Line number deletion successful end where it removes the //(double slash)

    Translating below text System.out.println("【○】COBOL Gives output System.out.println_○○ COBOL

    Where the special characters are removed and replaced with some other.

    I am reading my file using CP932 encoding and writing using UTF-8.

    Please let me know if there are any ideas on how to resolve this issue.

    I have tried Google translate and it works well on the same encoding.

    Stackoverflow link : https://stackoverflow.com/questions/57462276/translating-text-in-microsoft-translator-removes?noredirect=1#comment101605090_57462276

    1 vote
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)

    We’ll send you updates on this idea

    2 comments  ·  UserVoice  ·  Flag idea as inappropriate…  ·  Admin →
  17. Some cognitive service 'kind's only deploy if location="global"

    Services such as TextTranslate give an error "InvalidApiSetId - The account type 'TextTranslate' is either invalid or unavailable in given region." if the location is not set to "global".

    This is not mentioned in the documentation anywhere that I could find.

    1 vote
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Flag idea as inappropriate…  ·  Admin →
  18. QnA Maker formatting

    Our customers are finding difficult to deal with the special characters that QnA Maker shows to format the text in the answers (backslash+N, # for headers, links between [ ] and ( ), etc… ).
    Are there any plans to show and allow basic formatting?

    2 votes
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Other  ·  Flag idea as inappropriate…  ·  Admin →
  19. Change the font color of this web page

    The light color of blue used for text on this page is very difficult to read except in the very largest font sizes.

    2 votes
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Other  ·  Flag idea as inappropriate…  ·  Admin →
  20. Better handling of spoken numbers

    Currently the handling of spoken numbers in Speech to text is making it very annoying for my use-case.

    I am using LUIS to identify the following entities:

    {Group} {Flight} {Plane}

    Where Flight and Plane are individual single digit numbers. For example (Using the text representation of numbers when spoken to avoid confusion)

    "Enfield one two" spoken gets rendered to text and passed into LUIS as "Enfield 12" which round-tripping back to the text-to-speech is synthesized as "Enfield twelve". the lack of space between each digit also makes it impossible to assign them to individual entities (even using lists and regex…

    2 votes
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)

    We’ll send you updates on this idea

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

Feedback and Knowledge Base