Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/willianantunes/transcriber-wrapper
Wrapper of well-known transcribers that transform text into phoneme codes
https://github.com/willianantunes/transcriber-wrapper
arpabet espeak-ng festival-speech-synthesis international-phonetic-alphabet ipa linguistics mypy pytest transcriber transcription
Last synced: 3 months ago
JSON representation
Wrapper of well-known transcribers that transform text into phoneme codes
- Host: GitHub
- URL: https://github.com/willianantunes/transcriber-wrapper
- Owner: willianantunes
- License: gpl-3.0
- Created: 2021-02-27T18:36:06.000Z (over 3 years ago)
- Default Branch: main
- Last Pushed: 2021-04-08T14:40:09.000Z (over 3 years ago)
- Last Synced: 2024-07-16T10:53:52.693Z (4 months ago)
- Topics: arpabet, espeak-ng, festival-speech-synthesis, international-phonetic-alphabet, ipa, linguistics, mypy, pytest, transcriber, transcription
- Language: Python
- Homepage: https://pypi.org/project/transcriber-wrapper/
- Size: 99.6 KB
- Stars: 15
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Transcriber Wrapper
[](https://github.com/psf/black)
[](https://sonarcloud.io/dashboard?id=willianantunes_transcriber-wrapper)
[](https://sonarcloud.io/dashboard?id=willianantunes_transcriber-wrapper)
Inspired by [Phonemizer](https://github.com/bootphon/phonemizer), this a simpler version focused in transcription applications that work with IPA (International Phonetic Alphabet). This works like a wrapper which is responsible to call a back-end application, let's say [espeak-ng](https://github.com/espeak-ng/espeak-ng). It adds some features on top of it like `with stress` option.
## Supported back-ends
- [eSpeakNG](https://en.wikipedia.org/wiki/ESpeak)
- [Festival Speech Synthesis System](https://en.wikipedia.org/wiki/Festival_Speech_Synthesis_System)
## Usage
You need to install espeak-ng and festival on your operational system. See [Dockerfile.dev](./Dockerfile.dev) as an example. After that, you can create a transcriber and then use it in your logic:
```python
from typing import List
import transcriber_wrapper
# The standard language is "en-us"
# The standard back-end is "espeak"
transcriber_en_us = transcriber_wrapper.build_transcriber()
def do_the_thing(words: List[str]) -> List[str]:
return transcriber_en_us.transcribe(words)
```
Don't forget to see [test_builder.py](./tests/int/test_builder.py) to get insights how to use this project!
## Development
### Executing commands directly on the binaries
After building the remote interpreter service, just enter in it:
docker-compose run remote-interpreter sh
You must be at `/usr/bin/`. Then try one of these below.
#### eSpeakNG
Check out these links:
- [Supported Languages](https://github.com/espeak-ng/espeak-ng/blob/53915bf0a7cd48f90c4a38ac52fff697723d9f4d/docs/languages.md)
- [Command Line User Guide](https://github.com/espeak-ng/espeak-ng/blob/53915bf0a7cd48f90c4a38ac52fff697723d9f4d/src/espeak-ng.1.ronn)
Some sample commands:
```shell
espeak-ng "Hello my friend, stay awhile and listen!" -ven-us -x --ipa -q --sep=_
espeak-ng "Curiosity" -ven-us -x --ipa -q --sep=" "
espeak-ng "If you will not bow before a sultan, then you will cower before a sorcerer!" -ven-us -x --ipa -q
espeak-ng --voices
```
#### Festival
You can execute `festival --help` to get a list of what you can do through what the festival developers call "Shell API" (see more details [here](http://www.festvox.org/docs/manual-2.4.0/festival_28.html#Shell-API)).
You can use the script [festival.lisp](./scripts/festival.lisp) to get the computation from a given word, some samples:
```shell
WORD=something festival -b /app/scripts/festival.lisp
WORD=theoretically festival -b /app/scripts/festival.lisp
```
What you can do is just type `festival` and then start its command line prompt. From there you can do the following for example:
```shell
# It will list voices available
festival> (voice.list)
# Default voice
festival> voice_default
# This won't work with our Docker image, but if you are on your ubuntu/debian machine, it may will
(SayText "Can someone refactor festival to be writen in Python with a friendly API?")
```
### Updating pipenv dependencies
If you update Pipfile, you can issue the following command to refresh your lock file:
docker-compose run remote-interpreter pipenv update
If you'd like to add a new package, let's say a production one:
docker-compose run remote-interpreter pipenv install pyparsing
Don't forget to update your service!
docker-compose build remote-interpreter