feldberlin/timething

Timething is a library for aligning text transcripts with their audio recordings.

53
/ 100
Established

Leverages Wav2Vec CTC models from Hugging Face for forced alignment, producing character and word-level timestamps with confidence scores via the PyTorch forced alignment approach. Supports both long-form media (podcasts, audiobooks) and machine learning workflows, with optional audio re-cutting based on alignment results. Handles multiple audio formats and 12+ languages, running efficiently on CPUs or GPUs with configurable batch processing.

130 stars and 152 monthly downloads. No commits in the last 6 months. Available on PyPI.

Stale 6m
Maintenance 0 / 25
Adoption 15 / 25
Maturity 25 / 25
Community 13 / 25

How are scores calculated?

Stars

130

Forks

14

Language

Jupyter Notebook

License

MIT

Last pushed

Dec 03, 2024

Monthly downloads

152

Commits (30d)

0

Dependencies

8

Get this data via API

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

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