swiss-parliaments-corpus-paper

Code for the paper "Swiss Parliaments Corpus, an Automatically Aligned Swiss German Speech to Standard German Text Corpus", available here: https://arxiv.org/abs/2010.02810

Example

To run an example alignment of data/transcript.txt to data/audio.flac, follow these steps:

• Create a Python 3.7 environment
• Clone the repository
• Change your working directory to the repo directory
• pip install -r requirements.txt
• python -m spacy download de_core_news_sm
• python example.py

The output will be written to sentence_alignment.txt. You can compare it with the expected output (data/sentence_alignment_expected_output.txt) to make sure everything worked.