With MEETINGSTATION, you can convert the pre-recorded audio of meetings into character text.

In this transcript, there are no punctuation marks, no capital letters and, depending on the speakers and the noise and interference of the environment there are character errors. However, with good technical preparation, the accuracy of the transcript can be increased to over 90-95%, depending on the quality of the acoustic system and the environment. This depends on factors such as: the language spoken, the speakers' speaking ability, the quality of the microphones, ambient noise/noise cancellation, speech regeneration, etc.

The meeting transcript can be corrected, the audio recording and the transcript can be stored, transmitted and processed together. In applications where even a few errors in the transcript are corrected, there is no need for any manual editing. If a memo or transcript is written from the audio material, the speech recorder can reduce the transcriber's manual correction work by 66% (one third) on average.

Speech recognition on subscription, in real time or from an existing voice recording, works in online mode in over 100 languages, including automatic translation. Speech recognition is performed on servers abroad. For English, German, Spanish, French, German, Italian, Japanese and Chinese, speakers up to 6 can be automatically identified and tracked in the comments. Language identification can be used in three languages pre-selected from more than 100 languages.