revdotcom/fstalign
An efficient OpenFST-based tool for calculating WER and aligning two transcript sequences.
Built on OpenFST finite-state composition with lazy traversal algorithms, it aligns reference and hypothesis token sequences while supporting both NLP-formatted and CTM input formats for detailed error analysis. Version 2.0 introduces optimized graph traversal that dramatically accelerates alignment on lengthy or error-rich sequences, plus intelligent constraint handling that prevents punctuation-word misalignments and favors case-insensitive substitutions. Provides configurable beam search, strict punctuation enforcement, and JSON output for integration into speech recognition evaluation pipelines.
171 stars.
Stars
171
Forks
11
Language
C++
License
Apache-2.0
Category
Last pushed
Mar 12, 2026
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/voice-ai/revdotcom/fstalign"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
daanzu/kaldi-active-grammar
Python Kaldi speech recognition with grammars that can be set active/inactive dynamically at decode-time
nttcslab-sp/kaldiio
A pure python module for reading and writing kaldi ark files
kaldi-asr/kaldi
kaldi-asr/kaldi is the official location of the Kaldi project.
gooofy/py-kaldi-asr
Some simple wrappers around kaldi-asr intended to make using kaldi's (online) decoders as...
pykaldi/pykaldi
A Python wrapper for Kaldi