Audio and video data - transcription and annotation
Researchers prepare audio and video data in order to analyse interview content, annotate sounds or visual information, including the views and emotions of focus groups, or to take notes from video observations.
If you are preparing interviews, you will need to transcribe the audio to text.
If you are analysing the sounds from the audio or video, such as accents or pronounciations of words, you will need to make sure the sound file is clear and unblemished and create a clear set of annotations to describe the sounds.
Transcription
- Griffith Speech to text service : uses the Microsoft Azure tools to transcribe the data. It is available to all Griffith researchers, is suitable for use with sensitive data and has associated costs.
- Nvivo software student access staff access for automated transcription.
Watch this video for an overview on the Nvivo transcription feature.
- Alveo : Developed in Australia for researchers. The software breaks up audio into segments by identifying breathing pauses and provides a structured format for manual transcription. Griffith Education researchers have found it can reduce manual transcription time by up to 50%.
Watch the video below for instructions on how to use Alveo.
- YouTube or Vimeo : Both applications can generate text transcriptions of the audio and provide means and tools for corrections.
- Google docs type with your voice option and Android Live Transcribe can transcribe audio as it is spoken, it will work with voice to microphone but may not work with a recording.
- Microsoft Word dictation button : will transcribe as you speak, a way to get interviews transcribed as they are undertaken but will require corrections.
Warning: There are data security issues to consider when using the public data services above. The data may be shared with third parties, mined for information, or stored in an off-shore location.
Find out how to classify and treat your research data securely at Griffith’s Information Security Classification site.
Annotation
Annotation of sound files requires a different software type. If you are annotating a sound file you will need to be able to create markers on file to find, replay and make the required notes.
- Elan : downloadable software to annotate audio and video recordings. This tool is used by linguists to analyse sounds, accents and pronounciations. Supported by a number of international research centres including the (ARC) Centre of Excellence for the Dynamics of Language.
Next up : Process and analysis methods and tools.