The SpeechGraph software is a graph-theoretical analysis tool that uses text as input and graph features as output. SpeechGraph can run on many different platforms such as Linux, Windows and OSX.
Logo by G. M. Silva
We provide a User Guide as a PDF inside the software package, with a thorough explanation of all software functions. Alternatively, you can download it here.
You can download the latest version of the SpeechGraphs tool right here. The current sofware version is 1.0.
We would like to receive your suggestions or bug reports. Please email to email@example.com.
The SpeechGraphs software was developed at the Brain Institute of the Federal University of Rio Grande do Norte (Natal, Brazil), by R. Furtado, P.P.C., Maia, N.B. Mota, S. Ribeiro, M. Copelli and D.F. Slezak.
If you use SpeechGraphs, please cite:
1) Mota, N.B., Furtado, R., Maia, P.P.C., Copelli, M. & Ribeiro, S. Graph analysis of dream reports is especially informative about psychosis. Sci. Rep. 4, 3691; DOI:10.1038/srep03691 (2014)
2) Mota NB, et al. (2012) Speech Graphs Provide a Quantitative Measure of Thought Disorder in Psychosis. PLoS ONE 7(4): e34928. doi: 10.1371/journal.pone.0034928
In a recent study of our group (Mota, Pinheiro et al, unpublished data) historical text records have been used to assess characteristics of early language structure. Then, 448 texts from 3000 BC to 2010 were selected to compose main data set. 6 ancient traditions (classified as Syro-mesopotamian, Egyptian, Hinduist, Judeo-Christian, Greco-Roman and Persian), besides multicultural chosen from Medieval, Modern and Contemporary ages). Samples longer than 50 thousand words were trimmed to this size, reducing computational costs.
60 samples of poetry (from Medieval, Modern and Contemporary ages) were taken in order to analyze possible similarities with early cuneiform and hieroglyphic compositions.
Controlling arbitrary selection for latter texts, 200 sample texts were picked randomly using Random Page tool from Project Gutenberg, excluding poetry and plays. Texts from all steps of this investigation are standardized to English language.
You can download it here.