nttcslab-sp/kaldiio

A pure python module for reading and writing kaldi ark files

57
/ 100
Established

This tool helps speech researchers and developers easily manage the large audio feature and alignment files used in Kaldi speech recognition workflows. It reads and writes Kaldi's specialized 'ark' and 'scp' file formats, which contain matrices, vectors, and audio data. The end-user is typically an ASR researcher or engineer working with Kaldi.

268 stars. Used by 4 other packages. No commits in the last 6 months. Available on PyPI.

Use this if you need to programmatically read or write Kaldi's 'ark' and 'scp' data formats in a pure Python environment, without relying on C++ bindings.

Not ideal if you need to work with Kaldi's advanced neural network data formats (NNet2/NNet3 egs) or lattice files, as these are not supported.

speech-recognition Kaldi audio-features ASR-development data-management
Stale 6m
Maintenance 0 / 25
Adoption 14 / 25
Maturity 25 / 25
Community 18 / 25

How are scores calculated?

Stars

268

Forks

36

Language

Python

License

Last pushed

Mar 06, 2025

Commits (30d)

0

Dependencies

1

Reverse dependents

4

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/voice-ai/nttcslab-sp/kaldiio"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.