WeNet
Build speech-recognition training and inference workflows.
What’s included for free?
Ongoing-free Apache-2.0 source software for build speech-recognition training and inference workflows. No provider page, file, export, storage or execution quota is imposed by the software licence. Local CPU/GPU, RAM and disk constrain practical use; hosting, dependencies, input rights and optional services are separate. Model weights, training data and inference providers have separate terms and may require paid compute or API access.
Where it runs: Local install
Provider information ↗Checked 2026-11-05
Before you use it
Apache-2.0 for the core source. Preserve copyright, licence and required notices; review redistribution and source-sharing obligations where applicable. Bundled libraries, models, fonts and source documents have separate rights.
Provider terms can change. Check current pricing and usage rights before starting a project.