IO module

Input/output of data from files:

  • import EEG data from .set files or raw files

  • import features (word level features, envelopes, etc.)

Module to correctly extract Matlab files, especially EEGLAB .set files no matter if the version is >=7.3 and take the data to MNE format.

Another important part are classes to handle continuous stimuli aligned with EEG recording. These aligned features then serve as an input to most of the decoding functions.

Classes

AlignedSpeech(onset, srate[, path_audio, ...])

Generic class to describe features corresponding to a speech segment aligned with EEG data.

WordLevelFeatures([path_praat_env, ...])

Gather word-level linguistic features based on old and rigid files, generated from various sources (RNNLM, custom scripts for word frequencies, Forced-Alignment toolkit, ...)

Functions

eeglab2mne(*args, **kwargs)

extract_duration_audio(filepath)

Function to get the duration of an audio file

extract_duration_praat(fname)

Function to get the duration from a file generated by PrAat software, for instance to get envelope or pitch of a sound.

fieldtrip2mne(*args, **kwargs)

get_sentence_position(filepath)

Extract sentence position of words given a text file.

get_word_onsets(filepath)

Simply load word lists and respective onsets for a given sound file/story parts...

get_word_vectors(wordlist, wordvectors[, unk])

Get the word vectors for each word in the list supplied.

load_close_values(filepath)

Load syntactic structure open values.

load_depth_values(filepath)

Load syntactic structure depth values.

load_ica_matrices(fname)

Load ICA matrices from EEGLAB structure .set file

load_mat(fname)

Loading simple mat file into dictionnary

load_open_values(filepath)

Load syntactic structure open values.

load_surprisal_values(filepath[, eps])

Load surprisal values from the given file.

load_wordfreq_values(filepath[, key, ...])

Load word frequency, and returns -log of it (scaled by median value)