cmusphinx/pocketsphinx

A small speech recognizer

84
/ 100
Verified

Large vocabulary, speaker-independent continuous speech recognition using CMU's classical acoustic and language models optimized for compactness and efficiency on resource-constrained devices. Provides command-line tools and C/Python APIs for decoding audio files or streams, with capabilities including live speech detection, single-utterance recognition, and force-alignment of audio to text with word/phone/state-level segmentation. Outputs results as line-delimited JSON with confidence scores and temporal boundaries.

4,278 stars and 70,863 monthly downloads. Used by 2 other packages. Available on PyPI.

Maintenance 13 / 25
Adoption 22 / 25
Maturity 25 / 25
Community 24 / 25

How are scores calculated?

Stars

4,278

Forks

730

Language

C

License

Last pushed

Mar 09, 2026

Monthly downloads

70,863

Commits (30d)

0

Dependencies

1

Reverse dependents

2

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/voice-ai/cmusphinx/pocketsphinx"

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