Synote

Mike Wald is from the University of South Hamptom, has interests in Multimedia, Accessibility. He worked with the IBM speech recognition. Points made: speech is continuous but also needs to catch full stop, comas, other ponctuations and the common pauses in the speech.

  • Majority of speech recognition programs were founded by the American Defense Department.
  • Synote was created to help to include bookmarks and transcriptions into your audio/video.
  • Synmarks = bookmarks. They store only timings, notes and load the respective part of the audio.
  • Create synmarks highlighting the transcript. Each synmark has an unique URI.
  • Synote was created using the Google web toolkit (javascript + DHTML).
  • e-prints (eCS – South Hampton)

 

Another objective of DAMS

  • standard for interfaces (high level – XML)
  • data formats
  • object – XML metadata

 

Email instructions I received from M. Wald:

The current public version of Synote requires an XML file created with the speech recognition tool or by hand.

However you can try the new beta version of Synote which allows you to simply link to a video or audio without using an XML file:

http://synote-server.ecs.soton.ac.uk:8083/synote-r0425/

and I would welcome any comments you have.

We have not provided an updated guide for this yet but the new facilities include:

  • Synote also runs on Apple Mac (Recommend Safari Browser)
  • Synote can play Flash videos as well as recordings that play on media player
  • You can search the database of recordings, transcripts and Synmarks and replay the recordings found
  • Users can upload their multimedia: Select ‘create recording’ and fill in title, URL of where the recording is stored (Description and tags are optional)
  • You can create and edit slide/image URLs and timings (if you have the permission to do so)
  • You can create a synchronised transcript if you are the owner of the recording or have write permission. Select ‘edit’ and you can then load in a file if you want or if not you will either edit an existing transcript or create a new one. Select ‘next’ and you can now create or edit or synchronise text. Typing in the edit box at the bottom of the transcript window and then selecting ‘append text’ adds the text with the synchronisation point with the recordings current time. Selecting any word in the transcript moves the text between the synchronisation points to the edit box and allows you to edit the text and then select ‘change text’ to make the change. Hovering over a synchronisation point clock icon displays the synchronisation point time. Apart from the first and last ‘clocks’ they always appear in pairs to show the end time of the previous word and start time of the next. (These may be different when the transcript is created automatically using speech recognition but if created manually they will be the same).
  • You can insert a new synchronisation point pair of clocks with the current recording time by clicking in the space after a word.
  • You can change the time of a synchronisation point to the current recording time by selecting its clock icon. To delete a synchronisation point pair you must delete the text between this point and the next point. If you copy the text first you can paste it in after the previous word in the edit box so you don’t loose any text. The help button does nothing at the moment as we still have to write the help. Selecting ‘next’ allows you to preview the changes and ‘Finish’ saves the changes whereas ‘Exit’ does not save them. Selecting ‘back’ allows you to return to the previous stage. A transcript can have many synchronisation points and so saving these can take the browser a long time and so the browser may show a dialogue window telling you that a script is causing the system to be unresponsive. You must allow the script to run and NOT stop it and eventually it will save the changes (the time depends on the length of the recording and transcript and how many synchronisation points and so the dialogue window may come up many times). We are pushing what can be done with a browser using Javascript to the limits so please be patient.
  • You can assign permissions for logged in users to your recordings (‘private’ means only you can see it and change and annotate it and create and edit transcript and slides: ‘write’ means others can see it and create and edit, transcript , slides and create synmarks: ‘Annotate’means others can see it and create and edit synmarks) and you can also create ‘groups’ and give them different permissions than other logged in users.

Main link to the website: http://www.synote.org/synote/