techiaith-g2p
The permissive Bangor Welsh grapheme-to-phoneme (G2P) engine: turns Welsh (and code-switched English) text into the phoneme-id sequence consumed by Bangor University's retrained Piper/VITS voices. Dictionary-based, permissively licensed data only, and dependency-free -- it imports nothing beyond the Python standard library, so it drops into an inference server, a screen-reader add-on, or a CLI without pulling in a model runtime.
Install
pip install techiaith-g2p
Use
from techiaith.g2p import BangorG2P
g2p = BangorG2P(english_mode="native")
ids = g2p.text_to_ids("Bore da. Sut wyt ti'n teimlo heddiw?")
print(ids)
english_mode and the package's data_version() must match the target
voice's .onnx.json. For the full text-to-audio example, see the model card:
https://huggingface.co/techiaith/cy_en_GB-bu_tts
Do not feed this voice to stock piper. Its phoneme_type: "text" path
treats phonemes as individual codepoints, one at a time -- it does not
understand the multi-codepoint phoneme ids this G2P produces, and the output
will be mispronounced. Use the ids from text_to_ids directly against the
model's input layer, as the model card describes.
Links
- Voice model: https://huggingface.co/techiaith/cy_en_GB-bu_tts
- Pronunciation dictionary: https://github.com/techiaith/geiriadur-ynganu-bangor
- Source: https://github.com/techiaith/techiaith-g2p
- License: CC0-1.0 — public domain dedication (see
LICENSE; the bundled pronunciation data keeps its own BSD-2-Clause terms, seeNOTICE)
1 maintainer
Metadata
1 maintainer
Release files for techiaith-g2p 1.2.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| techiaith_g2p-1.2.0.tar.gz | 7.1 MB | Without provenance | Details for techiaith_g2p-1.2.0.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| techiaith_g2p-1.2.0-py3-none-any.whl | Python 3 | none | any | Details for techiaith_g2p-1.2.0-py3-none-any.whl |
Total release size: 13.0 MB
Release files / techiaith_g2p-1.2.0.tar.gz
| Download URL | techiaith_g2p-1.2.0.tar.gz |
|---|---|
| Size | 7.1 MB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
975198cdb58f4f21fe08d9aa63d60e136dd39d7443167df188436671bcaa2878
|
|
BLAKE2b-256 checksum How to use checksums |
deb775f8f7a3e38dd75fd8552d58d28bb9ec33390a35e95788404cddee52c584
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/7.0.0 CPython/3.10.21
|
Release files / techiaith_g2p-1.2.0-py3-none-any.whl
| Download URL | techiaith_g2p-1.2.0-py3-none-any.whl |
|---|---|
| Size | 5.9 MB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
3506b0ee9e06e4681d164a494b1d87cd0a2ed840b9782b44977525d1d36abf6d
|
|
BLAKE2b-256 checksum How to use checksums |
551487e7bbd15b58c86055bf8cc58c329ea64ef84af20616570bf6dd5275940d
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/7.0.0 CPython/3.10.21
|
Security for techiaith-g2p 1.2.0
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.
