Fig. 5
From: Free resources for forced phonetic alignment in Brazilian Portuguese based on Kaldi toolkit
The phonetic alignment pipeline followed by UFPAlign. When a user feeds the system with an audio (.wav) and its respective transcription (.txt), they should expect a TextGrid file as output. Time marks are provided by Kaldi, which relies on the knowledge of the acoustic model (AM) and tokens of the lexicon (L.fst). Preparation of data and language files (dashed block) are especially detailed in Fig. 6