Fact · Transcription
The 2017 Interspeech paper introduces the Montreal Forced Aligner as a new open-source system for aligning speech with text.
Checked against its source on October 17, 2026
In the source's own words
“We present the Montreal Forced Aligner (MFA), a new open-source system for speech-text alignment.”
Montreal Forced Aligner: Trainable Text-Speech Alignment Using Kaldi, Michael McAuliffe et al., Proceedings of Interspeech 2017, 2017, Abstract.
Open the original (DOI 10.21437/Interspeech.2017-1386)- What it measures
- The aligner takes a text transcript and a recording and times the words and phones in the recording. The paper compares its boundaries with human annotations on English conversational and laboratory speech.
- What it does not show
- The paper evaluates English speech only. Its accuracy comparison is against two older open-source aligners, not against neural alignment models built since 2017.