The first step in the process of developing substitutions is to prepare an efficient working environment. This environment contains the set of tools used to review transcripts and audio, develop substitution rules, and deploy the files to Voci ASR solutions. The following list shows the types of tools you will need and how you will use those tools.

Analytics tool

To efficiently locate and validate transcription errors, you will need an analytics tool capable of the following:

• Searching across large sets of transcripts

• Providing files that contain the search phrase

• Quickly navigating to the locations of a search phrase

• Simultaneously playing call audio and displaying the associated portion of the transcript

The analytics tool featured in this guide is V‑Spark, which is highly recommended for locating and validating transcription errors. V‑Spark offers powerful search capabilities including plain text, wildcard matching, and regular expression search options. Search results in V‑Spark direct you to associated call audio, which makes validating substitution candidates quick and efficient.

Python 2.7 interpreter

Voci provides Python scripts to assist with the process of developing substitutions. These scripts help you identify substitution candidates and check your substitution file for errors. You will need Python 2.7 installed on your computer to run these scripts.

Text editor

You will need a text editor to write a list of substitution rules. The text editor must be capable of saving in plain text. Freely available text editors such as Emacs, Vim, Nano, and Notepad++ work best as they save in plain text by default. If you prefer to use Microsoft Word, make sure to disable smart quotes.