Silero STT
Featherweight multilingual speech-to-text models you can load in two lines of PyTorch.
About
Silero STT is a set of compact pre-trained speech-to-text models from the snakers4/silero-models collection (6,100+ stars), loadable through torch.hub in a couple of lines. The repo's models.yml verifies coverage for English, German, Spanish and Ukrainian, with PyTorch, ONNX and TensorFlow SavedModel variants plus TF Hub listings, and the whole pitch is CPU-friendly speed for embedding transcription into applications. The catch is licensing: the repository LICENSE is CC BY-NC-SA 4.0, so commercial use is off the table without a separate arrangement — check it before shipping anything.
Specs
Context window
Pricing
Free quota
API access
Weights
Released
Benchmarks
No published independent benchmarks found; positioned as compact CPU-friendly models rather than accuracy leaders.
Provider-reported unless a source says otherwise. Benchmarks are marketing-adjacent by nature — treat them as directional.
Use cases
Trust score — 6/10
Scored 6/10 (verified tier, metrics-v1). Evidence: models.yml fetched and parsed (en/de/es/ua confirmed); PyTorch Hub listing verified; repo LICENSE fetched and confirmed CC BY-NC-SA 4.0. Strengths: genuinely simple to run, tiny footprint, multiple runtimes. Docked for: (1) non-commercial license blocks business use; (2) only four languages; (3) no published accuracy benchmarks. License: CC BY-NC-SA 4.0 (non-commercial).
How the trust score is built
Full methodology →- Adoption · stars/downloads, log-scaled vs peers
- 7.0/10
- Maintenance · recency, releases, issue load
- 6.0/10
- Openness · license + docs
- 4.0/10
- Manual review · sanity read + red-flag check
- 7.0/10
- · 6,134 stars on snakers4/silero-models
- · Languages verified in models.yml: en, de, es, ua
- · Ships PyTorch, ONNX and TensorFlow SavedModel variants
- · Two-line torch.hub loading; no hosted API