Knowledge Base

Export Voice Analysis Data

Aim

To export the expressions detected in your test participant's voice to a text file or The Observer XT format file.

Prerequisites

• You have the Voice Analysis Module.
• You carried out an analysis.

Procedure

1. Choose File > Settings and then click the Data Export tab.
2. Under Export (Detailed log, ODX, N-Linx and API), select Expressions from voice and then click OK.
3. Choose File > Export and select one of the options. See Export the data in FaceReader's output

The export file contains columns for each expression (Neutral from Voice, Happy from Voice, Sad from Voice and Angry from Voice), as well as for Valence (Valence from Voice), Arousal (Arousal from Voice), Loudness and Speech rate. Each expression has a value between 0 and 1, indicating the intensity of the expression. '0' means that the expression is not detected in the voice, '1' means that it is maximally present. Arousal from Voice, Loudness and Speech rate also have a value between 0 and 1. Valence from Voice has a value between +1 and -1. For more information see Voice Valence and Arousal Line Chart and Voice View.

The first milliseconds of the log file have 'Unknown' as value. This is due to the audio buffer which is used to collect approximately one second of data before starting the analysis.

Voice Expression Data In The Observer XT

If you send FaceReader data to The Observer XT 13 or higher versions using the N-Linx network communication protocol, Voice expression data are imported as external data. You can also manually import Voice expression data. To be able to import Voice expression data as external data you need the External Data Module of The Observer.

See FaceReader with The Observer XT

Source: FaceReader 10 Reference Manual (Help), Noldus Information Technology