| Title: | Write Spaces to Korean Sentences |
|---|---|
| Description: | Restores spacing in Korean sentences using a bundled neural network model and 'ONNX Runtime' <https://onnxruntime.ai/> through 'churon'. Inference runs locally without a web service or automatic downloads. |
| Authors: | Chanyub Park [aut, cre] (ORCID: <https://orcid.org/0000-0001-6474-2570>), Heewon Jeon [aut] (Author of KoSpacing code and bundled model assets) |
| Maintainer: | Chanyub Park <[email protected]> |
| License: | GPL (>= 3) |
| Version: | 0.1.5 |
| Built: | 2026-10-14 19:03:28 UTC |
| Source: | https://github.com/cran/theeuh |
Write spaces to Korean sentences.
space(ko_sents)space(ko_sents)
ko_sents |
A character vector of Korean sentences. Missing values are preserved. Sentences longer than 198 characters are truncated with a warning. |
The bundled model is loaded on the first non-empty, non-missing
sentence. ONNX Runtime must be configured first; see
churon::install_onnx_runtime(). No runtime is downloaded automatically.
On supported non-Windows runtimes, telemetry is disabled during model
initialization. If another package initializes ONNX Runtime first, set
ORT_DISABLE_TELEMETRY=1 before starting R to disable telemetry globally.
Sentences with restored spacing: a character string for one sentence,
or an unnamed list of character strings for multiple sentences. An empty
character vector returns list().
if (churon::check_onnx_runtime_available()) { space("\uC774\uC81C\uCEE4\uBC0B\uD558\uBA74") }if (churon::check_onnx_runtime_available()) { space("\uC774\uC81C\uCEE4\uBC0B\uD558\uBA74") }