Python wrapper for phonetisaurus grapheme to phoneme tool
Project description
Phonetisaurus for Python
Python wrapper for the excellent phonetisaurus grapheme to phoneme tool (license).
Includes pre-built binaries for:
x86_64
- desktop/laptop/server (64-bit)armv6l
- Raspberry Pi 0/1armv7l
- Raspberry Pi 2/3/4 (32-bit)aarch64
- Raspberry Pi 3/4 (64-bit)
Requirements
- Python 3.7+
- Linux
- Tested with Debian Buster
Installing
For x86_64
systems:
$ pip install phonetisaurus
For Raspberry Pi, see Releases for compatible wheels:
- Raspberry Pi 0/1
phonetisaurus-<VERSION>-py3-none-linux_armv6l.whl
- Raspberry Pi 2/3/4 (32-bit)
phonetisaurus-<VERSION>-py3-none-linux_armv7l.whl
- Raspberry Pi 3/4 (64-bit)
phonetisaurus-<VERSION>-py3-none-linux_aarch64.whl
Training
Assuming you have a lexicon formatted like the CMU pronouncing dictionary:
word1 phoneme1 phoneme2 ...
word2 phoneme1 phoneme2 phoneme3 ...
saved to lexicon.dict
run:
$ phonetisaurus train --model /path/to/write/g2p.fst /path/to/lexicon.dict
You may supply more than one lexicon.
See phonetisaurus train --help
for more options.
Predicting
$ phonetisaurus predict --model /path/to/g2p.fst word1 word2 ...
If no words are provided on the command line, they will be read line-by-line from standard in.
You may optionally supply one or more --lexicon /path/to/lexicon.dict
arguments to avoid guessing pronunciations for known words.
See phonetisaurus predict --help
for more options.
Project details
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distributions
Built Distribution
Hashes for phonetisaurus-0.3.0-py3-none-manylinux1_x86_64.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 4ad830d748234d778c9e55da731df319f92805120d158825cf8162616bbccc09 |
|
MD5 | 516d5e37a089098fa110bad0d37ba4cf |
|
BLAKE2b-256 | f8dd8bee1dc1f6944fec8a5a8a3a1e5a70e2680ef1e129c0ee946202a65ecf37 |