Skip to main content

The GreenKey ASRToolkit provides tools for automatic speech recognition (ASR) file conversion and corpora organization.

Project description

GreenKey Automatic Speech Recognition (ASR) Toolkit

The GreenKey ASRToolkit provides tools for file conversion and ASR corpora organization. These are intended to simplify the workflow for building, customizing, and analyzing ASR models, useful for scientists, engineers, and other technologists in speech recognition.


usage: convert_transcript [-h] input_file output_file

convert between text file formats

positional arguments:
  input_file   input file
  output_file  output file

optional arguments:
  -h, --help   show this help message and exit

This tool allows for easy conversion from STM files to TXT files and back. Other file formats will be added in the near future.


usage: wer [-h] [--char-level] [--ignore-nsns]
           reference_file transcript_file

Compares a reference and transcript file and calculates word error rate (WER)
between these two files

positional arguments:
  reference_file   reference "truth" file
  transcript_file  transcript possibly containing errors

optional arguments:
  -h, --help       show this help message and exit
  --char-level     calculate character error rate instead of word error rate
  --ignore-nsns    ignore non silence noises like um, uh, etc.


usage: [-h] files [files ...]

cleans input *.txt files and outputs *_cleaned.txt

positional arguments:
  files       list of input files

optional arguments:
  -h, --help  show this help message and exit

This script standardizes how abbreviations, numbers, and other formatted text is expressed so that ASR engines can easily use these files as training or testing data. Standardizing the formatting of output is essential for reproducible measurements of ASR accuracy.


usage: split_audio_file [-h] [--target-dir TARGET_DIR] audio_file transcript

Split an audio file using valid segments from a transcript file. For this
utility, transcript files must contain start/stop times.

positional arguments:
  audio_file            input audio file
  transcript            transcript

optional arguments:
  -h, --help            show this help message and exit
  --target-dir TARGET_DIR
                        Path to target directory


usage: prepare_audio_corpora [-h] [--target-dir TARGET_DIR]
                             corpora [corpora ...]

Copy and organize specified corpora into a target directory. Training,
testing, and development sets will be created automatically if not already

positional arguments:
  corpora               Name of one or more directories in directory this
                        script is run

optional arguments:
  -h, --help            show this help message and exit
  --target-dir TARGET_DIR
                        Path to target directory

This script scrapes a list of directories for paired STM and SPH files. If train, test, and dev folders are present, these labels are used for the output folder. By default, a target directory of 'input-data' will be created. Note that filenames with hyphens will be sanitized to underscores and that audio files will be forced to single channel, 16 kHz, signed PCM format. If two channels are present, only the first will be used.


usage: degrade_audio_file input_file1.wav input_file2.wav

Degrade audio files to 8 kHz format similar to G711 codec

This script reduces audio quality of input audio files so that acoustic models can learn features from telephony with the G711 codec.


usage: [-h] [--input-folder INPUT_FOLDER]
                                     [--output-corpus OUTPUT_CORPUS]

convert a folder of excel spreadsheets to a corpus of text files

optional arguments:
  -h, --help            show this help message and exit
  --input-folder INPUT_FOLDER
                        input folder of excel spreadsheets ending in .xls or
  --output-corpus OUTPUT_CORPUS
                        output folder for storing text corpus


  • Python 3.5 with pip


To ensure your dependencies install correctly, we recommend that you upgrade your setuptools before proceeding further.

pip install --upgrade setuptools

Now you are ready to install with pip.

pip install -e .


Code of Conduct

Please make sure you read and observe our [Code of Conduct].

Pull Request process

  1. Fork it
  2. Create your feature branch (git checkout -b feature/fooBar)
  3. Commit your changes (git commit -am 'Add some fooBar')
  4. Push to the branch (git push origin feature/fooBar)
  5. Create a new Pull Request


Original authors:

For all others who have aided this project, please see the [list of contributors].


This project is licensed under the Apache 2.0 License - see the file for details.

Project details

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

asrtoolkit-0.1.6.tar.gz (19.1 kB view hashes)

Uploaded source

Supported by

AWS AWS Cloud computing Datadog Datadog Monitoring Fastly Fastly CDN Google Google Object Storage and Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page