How can we make Microsoft Teams better?

← Public

Speech to text meeting notes

Using the technology of speech to text, have meetings record meeting notes in a text format based on what was spoken about in the meeting.

395 votes
Sign in
(thinking…)
Sign in with: Facebook Google
Forgot password?
Create a password
Signed in as (Sign out)
Close
Close

We’ll send you updates on this idea

Warren shared this idea  ·   ·  Flag idea as inappropriate…  ·  Admin →

47 comments

Sign in
(thinking…)
Sign in with: Facebook Google
Forgot password?
Create a password
Signed in as (Sign out)
Close
Close
Submitting...
An error occurred while saving the comment
  • Prasad commented  ·   ·  Flag as inappropriate

    It is a feature which can convert presenter's voice to text and display it as subtitles then and there on the video screen.

    1. some times it might not identify some business words, so there should be an option like user can add that word in to the grammar library, this option is open for all the attendees as well.

    2. option to high light the part of video and text segment which tells that the selected portion is portion is important.

    3.Give an option to speaker that he can make a note as in popup for that running video portion. so that it can be more descriptive and user can have an option to select his desired portion and play.

    4. Live recording video should have a option like episodes.
    Ex: if presenter is presenting on three topics , video should be chunked out as three episodes so that end user can directly go to his desired topic by clicking "next/prev" button.

    5. attendees should have an option like they can point/highlight the text/image which is presented by presenter.

    6. Add common white board concept, which can be accessible to all attendees as well.

    7. Speech to text concept should have localization feature.
    Ex: it shroud be able to convert the speech to user native language. for instance in India people are speaking in 14 languages, speech to text should be able to convert into these 14 languages.

    8. presenter should have a option to create MOM(Meeting of Minutes), and soon after completion of meeting he will be able to share the MOM using Mail.

    9. After completion of meeting speech to text should have an option to Export the converted text to Word/Notepad/Excel etc....

    10. Subtitle text should have a feature like auto correct spelling mistake.

    You can reach me on mail/contact number at any time for further clarification on the above features.

    Thanks
    Prasad DBR
    9591435112

  • Jonathan Freeman commented  ·   ·  Flag as inappropriate

    Once you go into a recording on MS Stream and set the language for it to English it will build an audio caption file that is searchable and can be downloaded. It has a long way to go but it is far better than nothing. Getting the file to at least recognize different speakers and tag them via their connection to the meeting would be a huge upgrade.

    AI to recognize individual voices in a board room would also be really cool but I'd place that behind just improving voice recognition in general. Presumably there is a learning process that is using the manual corrections made in streams by the users to basically crowd source the manual entry needed to help it improve accuracy.

  • Anonymous commented  ·   ·  Flag as inappropriate

    Alex

    We operate a two language staff meeting (Spanish & English) - we would love to see text of each language translated live as conversations occur in Teams meetings - this may be a stretch to far but if there is any way for this to be done pease advise me. Thank you

  • Anonymous commented  ·   ·  Flag as inappropriate

    Looks like this feature is live. After I recorded a meeting I received text notes from the people talking along with the recording. However, the notes was about 40% accurate

  • Anonymous commented  ·   ·  Flag as inappropriate

    Hello, a colleague recently used the Teams recording of a meeting to transcript. They were not using headsets. The transcript did not hear a word properly and transcribed it as a profanity. Any way for this Teams feature to prevent this?

  • Chris commented  ·   ·  Flag as inappropriate

    I had a demo of this in the MS offices in Reading last week. Looked great (and did auto translate too). Hoping it's out really soon?

  • Anonymous commented  ·   ·  Flag as inappropriate

    I would like to see an app built in to Microsoft Edge for a speech to text option. As I have cerebral palsy and it makes it very difficult to do all that typing. I used to use Dragon but I wish that Microsoft have something in their web browser to make it easier to fill out online forms.

    I'm a big fan of the edge browser.

    Matt

  • Catherine Pinto commented  ·   ·  Flag as inappropriate

    Hi I have a significant hearing loss and have a need for real time speech to text to understand offshore colleagues with strong Indian accents. My company uses Microsoft Skype for Business and Microsoft Teams. Can I enable it in either of these? It looks like you said above a feature is being worked on for teams - is it real time? Does it use AI to recognize accented voices? Also if it is not in teams, is it in available in skype for business? I am desperate for this feature to be able to be successful in my job. Thank you!

  • Miro commented  ·   ·  Flag as inappropriate

    Hi! Just to provide a bit of info & context I got with great help from Microsoft (Ana, Michael & Michael – thanks! :-) )

    - Functionality in essence does exist – MS Streams is accessible from MS Teams, and transcript functionality is part of MS Streams via Transcript Editor or via Captions file download.

    - With help of internal MS experts, Service Request #119012219575752 was placed through which I tried to ask for Transcript editor to be able to do SelectAll/PasteAll – but response was that Transcript editor was not designed for that purpose

    - As a separate request, there is also DCR # 28892 asking for transcript download (similar to captions file download, but without the need to filter out captions specific stuff but rather to get just recorded text)

    - Of course, captions file can easily be used today (with a bit of filtering/sorting – you can get clear text… tried it - works)

    - But – it seems major issue is audio quality and speech recognition engine itself (https://medius.studios.ms/Embed/Video/BRK3200?sid=BRK3200 ) . Basically, I did not succeed to get good enough quality of transcript that would serve as good enough transcription – or – that would serve at least as good enough basis for brief manual edit that can then produce good enough transcription … weird sentences, wrong words etc so it ended up rather me listening to recording one more time to write notes then to use/edit transcript

    - situations when people dial-in instead of using app/data connection, different conference rooms with different mic quality, people making noise etc etc seem to limit real usage currently to slow speaking non-dialect non-complicated meetings of 2-3 people all connecting from their devices without dial-in and via headphones/mic. Of course, it might be that I just did not do enough testing or did not do it in a right way

    So – looking forward to next MS Teams steps in making automated meeting notes taking really work!

  • noBob commented  ·   ·  Flag as inappropriate

    We are using the recorder in interviews and like to convert the recorded speech to text for further processing/editing.

← Previous 1 3

Forum Home

Reference Information for Teams

Public: Meetings

Categories

Feedback and Knowledge Base

(thinking…)

More Teams Information