LiGRE is equipped with a voice recognition system that can automatically transcribe the audio band of an audio or video interview featuring several persons, provided that they speak the same language. Uploading audio or video data into your qualitative project will automatically redirect you to the Automatic Transcription page. This page allows you to set the transcription parameters for your data and also provides you with an update of your automatic transcription features (for more detail, please refer to the screenshot below).
Once the transcription is completed, an email is sent to you to let you know that you can now start codifying it.
* Refer to the Add Data page for information on the audio and video formats that can be transcribed by LiGRE.
If, for some reason, you cannot launch the automatic transcription after uploading a file, know that you can always do it from the Interactive Report of the Codification and Data page.
Find the row featuring the file that you want to transcribe, look for the Transcribe Automatically option in the Actions column and click on it. It will bring you to the Automatic Transcription page.
LiGRE’s automatic transcription tool can transcribe over forty different languages. Here’s the list:
* LiGRE allows you to transcribe and codify data in right-to-left languages like Arabic and Hebrew.
The quality of the transcription depends on the quality of the speech recorded and of the audio band. As such, it is better to import a file that:
- has been recorded with a microphone (preferably unidirectional) that is well positioned –No need to invest a fortune in the purchase of a microphone, any “pick-up” microphone for smartphones will yield results that will be more than satisfactory.
- has a clean audio stream– Do not try to improve the sound quality of your audio and video data (e.g., noise reduction) artificially before importing it into the application since this type of operation tends to reduce the accuracy of the recognition system. This service has been designed to handle light ambient noise.
If you are making group recordings, the recognition system will produce better results if you add silences of a few seconds between each intervention.
Automatic Transcription Fees
Automatic Transcription is billed per second but bought by the hour. The duration of your interview determines both the amount of automatic transcription time used and the price of the transcript.
Automatic transcription time works independently of your project, which means that you keep all your remaining transcription time when you close your current project and, also, that you can use it in another project.
* Note that the transcription will not launch unless you have enough automatic transcription time to cover the length of your interview. You can always buy time from the Application menu or from your Project Dashboard.
Automatic Transcription and Team Project
The Automatic Transcription feature can be used in a team project, just like in a single user project. That said, only the automatic transcription time of the project owner will be used, which means that he/she must provide the transcription time for the entire team.
* It is important to note that the team members can only use the time of the project owner while working on his/her project.