MFA Input Preparation

[HamidMojarrad]

Hello everyone,

I am currently working on a corpus that includes a substantial number of transcript and audio files in .txt and .wav formats. Each interview in the corpus also has an associated .csv file containing two columns: Speaker (indicating the speaker ID) and Text (representing the corresponding utterance).
Before conducting my research, I need to align these files using MFA. However, to reduce alignment time, I aim to first convert the data into TextGrid format with two tiers, one for each speaker to make a standard input for MFA.

Could you advise on the best method to automate this conversion process? A manual approach would be impractical given the size of the dataset. I really appreciate any effective recommendations.