MontrealCorpusTools/Montreal-Forced-Aligner

Health v2 · maintenance only

Adoption not part of the score

1872 stars · 285 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

Montreal Forced Aligner is a command line utility for time-aligning orthographic transcriptions and pronunciation dictionary entries to audio recordings, built on the Kaldi ASR toolkit. It ships with pre-trained acoustic models and supports training custom models and grapheme-to-phoneme dictionaries for any language.

Use cases

• align transcriptions to audio recordings of speech corpora
• generate word and phone-level timestamps for speech data
• train acoustic models for a new language
• generate pronunciation dictionaries with grapheme-to-phoneme conversion
• segment long recordings into utterances for phonetic analysis

When to choose

• you need accurate word/phone-level alignment for phonetic or linguistic research
• you have a pronunciation dictionary and transcribed audio in any language
• you want a scriptable CLI with pre-trained models instead of building Kaldi pipelines yourself

When to avoid

• you need real-time streaming speech recognition rather than offline alignment
• you have no transcriptions or pronunciation dictionary for your audio
• you need a GUI-based annotation tool

Facets

cli-tool · maturity active

speech-recognition nlp cli machine-learning speech-processing developer-tools windows python cli forced-alignment kaldi g2p phonetics acoustic-models pronunciation-dictionaries speech-corpora natural-language-processing linux macos

Member repositories

For agents

markdown · JSON · MCP: product_card(name="MontrealCorpusTools/Montreal-Forced-Aligner")