Skip to content

Phonemizer

The phonemizer allows simple phonemisation of words and texts in many languages. It is based on four backends: espeak, espeak-mbrola, festival and segments.

Phonemizer is available as an Apptainer container on Apocrita.

Usage

To run the default installed version of Phonemizer, simply load the phonemizer module:

$ module load phonemizer
$ phonemizer --version
phonemizer-VERSION
available backends: espeak-ng, espeak-mbrola, festival, segments

For full usage documentation, run phonemizer --help.

The phonemizer Pip package has been installed inside the Phonemizer container, which can be used by launching the python program after loading the phonemizer module, as shown below:

$ module load phonemizer
$ python
>>> import phonemizer
>>> phonemizer.__version__
'X.Y.Z'

Example jobs

Serial jobs

Here is an example job running running on 1 core and 1GB memory:

#!/bin/bash
#SBATCH -n 1               # (or --ntasks=1) Request 1 core
#SBATCH --mem-per-cpu=1G   # Request 1GB RAM per core
#SBATCH -t 1:0:0           # Request 1 hour runtime

module load phonemizer

# Phonemize phrase using English (GB) dialect within the espeak backend engine
echo 'The quick brown fox jumps over the lazy dog' \
  | phonemizer -l en-gb -b espeak

Here is an output this job will produce:

ðə kwɪk bɹaʊn fɒks dʒʌmps əʊvə ðə leɪzi dɒɡ

Here is an example job running running using the phonemizer Python package inside the container, on 1 core and 1GB memory:

#!/bin/bash
#SBATCH -n 1               # (or --ntasks=1) Request 1 core
#SBATCH --mem-per-cpu=1G   # Request 1GB RAM per core
#SBATCH -t 1:0:0           # Request 1 hour runtime

module load phonemizer

python -c "from phonemizer import phonemize; \
        print(phonemize(['line1', 'line2']))"

Here is an output this job will produce:

['laɪn wʌn ', 'laɪn tuː ']

Reference