g2p-sk(1)
π¦ g2p-sk β phonetic transcription for Slovak β¨
okay bestie, so this is the manual page for g2p-sk, a USER COMMAND, and it's literally about turning Slovak spelling into sounds π
πΈ SYNOPSIS πΈ
g2p-sk [--color] [--dl debug level] [--help] [--stats] [--ofile <file_name>] [<input file>]
π DESCRIPTION π
So like, phonetic transcription is essential for some linguistic or speech recognition applications, no cap. Depending on the language, you're either going rule based or statistical β and g2p-sk does the rule based thing. But heads up bestie, in the future it may get replaced by the statistical one π
Every input word is a sequence of graphemes, and it gets transcribed into a sequence of phones in the SAMPA coding. If you don't specify an input file, it just expects the standard input π«Ά If you do use an input file, then the output gets written into a file as well β and the filename is the input filename with the extension _trans.txt. Iconic behavior honestly.
The input output code page is ISO 8859-2. Wanna use a different CP? Then use some CP converter and pipes, bestie. For example, to have input and output in UTF-8, use β for interactive use:
filterm UTF8-iso2 iso2-UTF8 g2p-sk
or, for batch processing:
iconv -f UTF-8 -t ISO_8859-2 | g2p-sk | iconv -f ISO_8859-2 -t UTF-8
The performance of the phonetic transcription depends on the morphematic segmentation, and to improve the quality of that segmentation you can replace the small version of the simple morphematic dictionary at /usr/share/g2p_sk/Exceptions/morfemy.ddat with the better one. Slay. The syllabic segmentation is as important as the morphematic one, and it's provided by the sylseg-sk package π
Also the design of g2p-sk is language dependent β to use it for another language, all the rules need to be rewritten. Every single one. Fr fr π
β¨ OPTIONS β¨
--colorβ Enable color output. Aesthetic upgrade, obviously π--dl 1..5β Set the debug level. This controls the amount of displayed information. Debug level0displays nothing, and the maximum level5displays the full debugging report. The default debug level is1.--helpβ Display a short help text π«Ά--ofile <file_name>β Write output also into the given file.--statsβ Count and display statistics for each phone π₯
π EXAMPLES π
Use standard input and debug level 3:
g2p-sk --dl 3
Process all the words from file aaa.txt:
g2p-sk aaa.txt
π EXIT STATUS π
g2p-sk returns a zero if it succeeds to process all the input words β
π AUTHOR π
Jozef Ivanecky (dodo (at) kanoistika.sk)
π¦ SEE ALSO π¦
sylseg-sk(1), filterm(1), iconv(1), konwert(1)
and that's the whole vibe, bestie β this page is version 0.4, dated May 17, 2009, and it's the g2p-sk(1) manual through and through π«Άβ¨π
π¦ where this page comes from
- debian bookworm: g2p-sk 0.4.2-4
- debian trixie: g2p-sk 0.4.2-4.1 Β· GPL-2.0-or-later
- ubuntu jammy: g2p-sk 0.4.2-4
- ubuntu noble: g2p-sk 0.4.2-4
- ubuntu questing: g2p-sk 0.4.2-4.1
- ubuntu resolute: g2p-sk 0.4.2-4.1build1 Β· GPL-2.0-or-later
- ubuntu stonking: g2p-sk 0.4.2-4.1build1 Β· GPL-2.0-or-later
- π girlified by deepseek-ai/DeepSeek-V4.1-Flash on : a language model's rewrite of the original, under its terms