Microsoft

Speech Service

                               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.



  1. enable shorter keywords

    Even though they might be less accurate, please enable the possibility to train models using a short keyword for custom keywords. For specific applications this would be extremely useful.

    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  ·  Custom Speech  ·  Flag idea as inappropriate…  ·  Admin →
  2. enable shorter keywords

    Even though they might be less accurate, please enable the possibility to train models using a short keyword for custom keywords. For specific applications this would be extremely useful.

    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  ·  Custom Speech  ·  Flag idea as inappropriate…  ·  Admin →
  3. Masking Sensitive Information

    Dealing with Numbers
    After transcribing audio into text, we have a Python script were we are trying to deal with privacy. Therefore, we are masking all the numbers (e.g. 0,1,2,3,4,5,6... and one, two, three etc.).

    Unfortunately, this is not fulfilling, since we want to remove on the one hand all the personal info. However, on the other hand keeping product information (e.g. serial number, product number etc.).

    Therefore my question: are there any best practices for this problem (since I cannot imagine that we are the first one dealing with this :-)).

    Many thanks in advance!

    Example:
    Original: Hi, my…

    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  ·  Speech to Text  ·  Flag idea as inappropriate…  ·  Admin →
  4. speech studio

    I have been trying to test out the Audio Content Creation. But this error message pops up: "The server is currently unable to handle your request", every time I try. Please advise

    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  ·  Text to Speech  ·  Flag idea as inappropriate…  ·  Admin →
  5. German: einbauen

    Pronounciation of german "einbauen" is terrible. The rest is almost impossible to tell computers and humans apart.

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

    We’ll send you updates on this idea

    1 comment  ·  Text to Speech  ·  Flag idea as inappropriate…  ·  Admin →
  6. C# Examples for Speech SDK

    It would be nice if the SDK documentation and GitHub repo included examples in C# and not just Java and C++.

    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 →
  7. Welsh....

    Is Welsh anywhere on the road map? In the UK there are some legal requirements for services to be available in Welsh alongside English.

    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  ·  Text to Speech  ·  Flag idea as inappropriate…  ·  Admin →
  8. set time out for batch transcription requests

    From what I see in the documentation, there's no option to set a timeout for a batch transcription job. I would like the transcription that is in "NotStarted" status for more than a specific amount of time to be discarded. And I would like to set the timeout value while I am creating the request

    I am doing batch transcription in a loop for a set of audio files one after another. For one of the transcription jobs which was created on 1:26 pm EST 10/22/2020 and until now (2:46 pm) is still in "NotStarted" state. I am not even…

    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  ·  Speech to Text  ·  Flag idea as inappropriate…  ·  Admin →
  9. Any CPU build

    Currently the .NET project has to be built in either x64 or x86. Any CPU is not supported. Please change this

    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  ·  Speech to Text  ·  Flag idea as inappropriate…  ·  Admin →
  10. Don't pronounce initials as abbreviations in Dutch

    Various initials are mispronounced as abbreviations in Dutch by text-to-speech. Since Teams uses this service for the voicemail messages, this is very annoying. These are a few examples:

    b. - pronounced as bruto
    c. - pronounced as cent
    e. - pronounced as edidiet
    f. - pronounced as forte
    H. - pronounced as heilige
    i. - pronounced as in
    l. - pronounced as loco
    p. - pronounced as piano
    t. - pronounced as temperatuur
    H.M. - pronounced as Hare Majesteit (Her Majesty)
    V.J. - pronounced as vorig jaar (last year).
    G.B. - Gedaan en bieden
    B.G. - Bovengenoemde

    We would like…

    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  ·  Text to Speech  ·  Flag idea as inappropriate…  ·  Admin →
  11. integration in Microsoft teams

    integration in Microsoft teams

    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  ·  Speech Translation  ·  Flag idea as inappropriate…  ·  Admin →
  12. 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  ·  Custom Voice  ·  Flag idea as inappropriate…  ·  Admin →
  13. 0 votes
    Sign in
    (thinking…)
    Sign in with: Facebook Google
    Signed in as (Sign out)

    We’ll send you updates on this idea

    0 comments  ·  Custom Voice  ·  Flag idea as inappropriate…  ·  Admin →
  14. Neural voices in Northern Europe

    Make neural voices available in the Northern Europe region.
    We want to use the Norwegian Neural voice in our product (based in Norway, naturally) but are for some reason locked out of this feature.

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

    We’ll send you updates on this idea

    0 comments  ·  Text to Speech  ·  Flag idea as inappropriate…  ·  Admin →
  15. Rest API to accept authorization token for RecordingsUrl

    Hi, my company has been using Azure's speech to text service and are happy with the results.

    However, we've hit a snag. For the RecordingsUrl parameter, I understand we will be passing in a blob uri that's public facing (no auth required), but our audio files are stored in a 3rd party service whereby an auth bearer token is required to be passed into the header to be able to access the .wav file (request header: { Authorization: Bearer <token> }).

    For the speech to text Rest API, is it possible if we can have an additional parameter (eg. RecordingsUrlToken)…

    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  ·  Speech to Text  ·  Flag idea as inappropriate…  ·  Admin →
  16. Custom Voice Portal does not have an option to add tests

    The portal says "Add a test", but there is no option to do so. Screenshot attached

    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  ·  Custom Voice  ·  Flag idea as inappropriate…  ·  Admin →
  17. Améliorer la lecture des nombres pour la Suisse

    Bonjour,
    La voix fr-CH, French (Switzerland), Male, "fr-CH-Guillaume" ne lit pas les nombres comme nous le faisons en Suisse, elle le fait comme en France.
    En effet, 70 doit se dire "septante" et non "soixante-" et 90 doit se dire "nonante" et non "quatre-vingt-". En outre selon les régions 80 se dit "huitante". Ce qui en rapport à la majorité des autres langues devrait être la norme.
    Avec mes meilleures salutations,

    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  ·  Text to Speech  ·  Flag idea as inappropriate…  ·  Admin →
  18. Improve workflow for Intent recognition training

    I used the following workflow for training my intent recognition:
    1) I've a series of entities, features and patterns edited
    2) I've a series of example inputs for training
    3) all the samples have the entities marked
    4) now I train the examples
    5) execute a series of batch test cases

    The issues recommended for improvement
    - The test cases for the batch testing require character positions, startPos and endPos. I had no other option than counting these manually, which is error prone.
    - When loading the batch test cases, the feedback / error log is hard to find and…

    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  ·  Speech to Text  ·  Flag idea as inappropriate…  ·  Admin →
  19. ccextractor

    Hello,

    May I recommend you use the closed-captioned text extractor tool called, "ccextractor" in order to compare the results of your Speech-to-Text service.

    The url is:

    https://www.ccextractor.org/

    Thank you.

    Regards,
    William Johnson

    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  ·  Speech to Text  ·  Flag idea as inappropriate…  ·  Admin →
  20. Support for specifying an External ID when creating a batch transcription request, which will be part of the response of the Web Hook

    It would be nice if you could specify an “extern id” when you create a batch transcription request and that the “external id” is also returned in the response of the web hook callback.

    Why? To be able to link a request to an id of a running process/workflow. For example in a durable function. The durable function (using a orchestration) looks like:
    1. Durable function send a message to the speech to text service (STTS) to create a transcription.
    2. Durable function makes a call to context.WaitForExternalEvent<string>("TranscriptionCompleted");
    3. At some point in time the STTS is finished and calls…

    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  ·  Speech to Text  ·  Flag idea as inappropriate…  ·  Admin →
← Previous 1 3 4 5
  • Don't see your idea?

Feedback and Knowledge Base