Speech2txt Manual - selmling/Analytics-and-Data-Exploration GitHub Wiki

Goal of Task

the purpose of the this script is to automate speech to text transcription . The goal of this project was to utilize existing technologies and scripting tools to automate the transformation of raw audio files into well formatted text files with useful information such as timestamps. Running the scripts properly, with the proper libraries and software packages installed should result in a .txt file formatted in the same way as the lab's speech transcription files manually derived from speech transcription in Audacity.

Requirements

Before running the script, you need to make sure that you have the following requirements installed on your system:

  1. Python: Anaconda distribution has a lot of important packaged (such as NumPy and Pandas) already installed and it comes with the conda package manager as well.

  2. pip: If you choose to get a distribution of python other than Anaconda, you will need to install the python package management system that is pip.

  3. Praat: This is the software that is needed to get the portions of sound and silence from a raw audio file. It will be needed to run the supplementary script mentioned earlier.

  • Python Packages: to run the script, you will need to install the following Python packages. You can install these using either the conda or pip package managers.

    1. pydub

    2. SpeechRecognition

    3. pandas

    4. os

    5. parselmouth

Text Editor: While basic text editors such as TextEdit(macOS) and Notepad(Windows) would work perfectly fine, installing a more visually advanced text editor to make running the script easier - Atom - is recommended.