Recording

The host clicks the Create/Join Meeting button on the M>D site and may schedule a meeting; having a name, planned date and time, and the language that will be spoken. The M>D server will produce a link and a meetingID+password. The host will send these meeting details by Email to all participants as an invitation.

The recording is started/paused/ended by the host of the meeting. While pausing the recording the video conference will continue uninterrupted, thus allowing some off the record discussions. Participants are requested to sit in a not too noisy environment, for optimal microphone performance.

When the meeting is ended all participants will get the message that audio is sent to the M>D server. Do not switch-off the laptop before you have the confirmation that all recordings are transferred. Your data is only used for transcribtion purposes.

AutoTranscribe

All your hosted meetings are found in the M>D Portal. You may also assign the role of editing to someone else with a M>D account.
Before starting the transcription process the system will calculate the cost of the AutoTranscription, including the use of the M>D workbench: a set of tools for editing the AutoTranscript and compiling the minutes of the meeting. At this point the user may also invoke the vocabulary function: the AutoTranscription function is fed with a phrase-set of specific words in the context of the specific meeting. This will greatly enhance the accuracy of the AutoTranscript.

For larger organizations M>D has the central billing feature. All authorized employees can book the costs of M>D on the central account. Please contact us for consolidated billing every the end of the month.
During the AutoTranscription process, the individual recordings per user are fed into the Speech-to-Text-Engine (StTE). Because each feed contains only one voice – or voice profile – the StTE can optimally adapt to this specific voice profile, resulting in the best possible AutoTranscript.
The sound bites are processed into text fragments. These text fragments are reassembled on a time line into one document: the AutoTranscript of the meeting.

Editing

The M>D AutoTranscript is very good, but not without errors. If necessary, the AutoTranscript may be corrected while listening to the original audio.
The editing screen comprises of the original AutoTranscript on the left, the Edited Transcript on the right, and an audio bar on top. These three parts of the editing screen are synchronized. The user may click anywhere on the audio bar or the AutoTranscript and the recording will play form there. The cursors in both text fields and the audio bar are moving in sync while playing.
When clicking on the Edited Transcript, audio play will pause and corrections can be made. After the correction, audio play may be resumed by a right-hand mouseclick, clinking on the audio bar or AutoTranscript.

The editor may highlight parts of the text (maybe any color) in the righthand Edited Transcript part of the screen. Highlighted text will be automatically transferred to the Summary screen. This highlighted text fragments are the building blocks for the minutes of the meeting.
Note: in many cases there’s no need for a perfect Transcript. The original AutoTranscript will do and is used only to compile the summary or minutes of the meeting, by means of the handy and fast copy/paste functionality, highlighting the most relevant parts of the AutoTranscript.

Summary

When opening the Summery screen, the editor will find again two text parts and the audio bar, now the EditedTranscript on the left and the Summary on the right. The previously highlighted text fragments are tranferred to the Summary, labeled with speaker and ordened by timestamp. These text fragments are the building blocks of the summary or minutes of the meeting. The editor may extend and polish the Summary by using the Fast Copy/Paste function, possibly while listening to the audio. 

The editor may add extra text to bind the text fragments from the Transcription into one consistent document and… done… the minutes of the meeting are compiled based upon the actual spoken text fragments, with great accuracy, looking smart, but with a minimum effort.