metadata
license: bsd-3-clause
language: th
inference: false
tags:
- secryst
- transliteration
- grapheme-to-phoneme
- onnx
- byt5
tha-g2p-small-1.0
Part of the Secryst phonological-layer family — a dedicated model for g2p (TH), distributed as a Secryst IMF artifact and loadable through any Secryst crystal:
from secryst import Model
Model.load("tha-g2p-small-1.0")
- Artifact:
tha-g2p-small-1.0-int8.zip(0.24 GiB) - Precision variants:
tha-g2p-small-1.0-int4
Measured quality
Full evaluation sets, zero skipped examples, the benchmark's own evaluator. Provenance for every number is the results record of the training project.
| metric | value | provenance |
|---|---|---|
| per_teacher | 4.43 | interscript/interscript-ml docs/RESULTS.md#tha-g2p-small-1.0 |
| per_student_beam4 | 12.06 | interscript/interscript-ml docs/RESULTS.md#tha-g2p-small-1.0 |
| per_student | 2.85 | interscript/interscript-ml docs/RESULTS.md#tha-g2p-small-1.0 |
License
BSD-3-CLAUSE
Provenance
Trained in the Secryst training monorepo and published through the interscript-ml distribution contract (models.yaml index, sha256-verified release assets, parity-gated artifacts). The canonical download channel is GitHub Releases; this repository is the Hugging Face mirror of the same bytes, pinned to the index revision.