Skip to content

Code for the paper "A high-performance speech neuroprosthesis"

Notifications You must be signed in to change notification settings

anna-charlotte/speechBCI

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

58 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

A high-performance speech neuroprosthesis

System diagram

Overview

This repo is associated with this paper, dataset and machine learning competition. The code contains the RNN decoder (NeuralDecoder) and language model decoder (LanguageModelDecoder) used in the paper, and can be used to reproduce the core offline decoding results.

The jupyter notebooks in AnalysisExamples show how to prepare the data for decoder training, train the RNN decoder, and evaluate it using the language model. Intermediate results from these steps (.tfrecord files for training, RNN weights from my original run of this code) as well as the 3-gram and 5-gram language models we used are available here (in the languageModel.tar.gz, languageModel_5gram.tar.gz, and derived.tar.gz files).

Example neural tuning analyses (e.g., classification, PSTHs) are also included in the AnalysisExamples folder (classificationWindowSlide.ipynb, examplePSTH.ipynb,naiveBayesClassification.ipynb, tuningHeatmaps.ipynb, exampleSaliencyMaps.ipynb).

PyTorch Implementation

We have posted a new PyTorch implementation of this original TensorFlow RNN decoder here.

Train/Test/CompetitionHoldOut Partitions

We have partitioned the data into a "train", "test" and "competitionHoldOut" partition (the partitioned data can be downloaded here as competitionData.tar.gz and has been formatted for machine learning). "test" contains the last block of each day (40 sentences), "competitionHoldOut" contains the first two (80 sentences), and "train" contains the rest.

The transcriptions for the "competitionHoldOut" partition are redacted. See baselineCompetitionSubmission.txt for an example submission file for the competition (which is generated by this notebook).

Results

When trained on the "train" partition and evaluated on the "test" partition, our original run of the code achieved an 18.8% word error rate (RNN + 3-gram baseline) and a 13.7% word error rate (RNN + 5-gram + OPT baseline). For results on the "competitionHoldOut" partition, see the baseline word error rates here.

Installation

NeuralDecoder should be installed as a python package (pip install -e .) using Python 3.9. LanguageModelDecoder needs to be compiled first and then installed as a python package (see LanguageModelDecoder/README.md).

About

Code for the paper "A high-performance speech neuroprosthesis"

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Roff 48.3%
  • Jupyter Notebook 22.9%
  • C++ 18.3%
  • Python 3.5%
  • C 2.5%
  • HTML 1.7%
  • Other 2.8%