What is FAVE?

screen shot of file transcribed in ELAN


HMM symbol

Forced alignment

screen shot of wave form and spectrogram with vowel formants

Vowel extraction

Plotnik screen shot

Vowel analysis

The FAVE (Forced Alignment & Vowel Extraction) program suite allows you to automatically align and extract large quantities of vowel formant measurements from sociolinguistic interviews or other bodies of orthographically transcribed data.

FAVE is a set of two programs: FAVE-align and FAVE-extract.

  • FAVE-align

    FAVE-align is a forced alignment program adapted for sociolinguistic interviews or other texts with multiple speakers. It accepts as input a sound file with its corresponding orthographic transcript, and returns a Praat TextGrid file with two tiers per speaker, a phone tier and a word tier.

  • FAVE-extract

    FAVE-extract automatically measures the formant values for F1 and F2 for all vowels for a given speaker. Its input is a sound file, plus a corresponding aligned TextGrid with word and phone tiers for each speaker (typically the output of FAVE-align).

How does it work?

The FAVE-align and FAVE-extract interfaces allow you to upload your files to the server, which then processes them in the background, and will send you the resulting files as an email attachment.

Please note that depending on the volume of traffic for the web site, the processing of your uploaded files may take a while. Do not expect immediate results! You can, however, generally expect to receive your results within a day.

Detailed instructions on how to use FAVE-align and FAVE-extract can be found on these pages:

Can I use FAVE with languages other than English?

At present, FAVE works with English-language data only.