https://github.com/pythainlp/pipeline
https://github.com/pythainlp/pipeline
Last synced: about 1 month ago
JSON representation
- Host: GitHub
- URL: https://github.com/pythainlp/pipeline
- Owner: PyThaiNLP
- License: apache-2.0
- Created: 2024-07-09T14:42:28.000Z (over 1 year ago)
- Default Branch: main
- Last Pushed: 2024-07-09T14:55:32.000Z (over 1 year ago)
- Last Synced: 2025-03-05T10:48:05.433Z (about 1 year ago)
- Homepage: https://pythainlp.org/pipeline/
- Size: 6.84 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Open Source Pipeline
Do you want to do someting about Thai language from Open Source? Here is the guideline about resource for working with Thai language in your open-source AI project!
## Smart Speakers
For smart speakers task, you want to do:
- Speech-to-text model
- Keyword Spotting (KWS)
- NLU
- Text-to-speech model
### Speech-to-text model
You can use openai whisper mdoel or other model that fine-tuning pretrained model to Thai language. If you want to train your model, you can download Thai speech corpus from [nlpforthai.com/tasks/speech-recognition/](https://nlpforthai.com/tasks/speech-recognition/).
### Keyword Spotting
Thai language have [Gowajee corpus](https://github.com/ekapolc/gowajee_corpus) for Thai smart home by "Gowajee" hotword.
### Text-to-speech
You can use Text-to-speech from [The Massively Multilingual Speech (MMS) project](https://github.com/facebookresearch/fairseq/tree/main/examples/mms) or use Thai open-source model from [nlpforthai.com/tasks/speech-synthesis/](https://nlpforthai.com/tasks/speech-synthesis/).
> MeloTTS are working to support Thai. [https://github.com/myshell-ai/MeloTTS/issues/120](https://github.com/myshell-ai/MeloTTS/issues/120)