---
license: cc-by-nc-sa-4.0

language:
- lb
- de
- fr
- en
- pt

tags:
- STT
- ASR
- audio
- speech recognition
- coqui.ai

datasets:
- mbarnig/lb-STT-CORPUS

---
#### The luxembourgish part of my multilingual automatic speech recognition (ASR) model is the second Machine Learning (ML) model for Luxembourgish. The very first model has been published in May 2022 by [Pr Peter Gilles](https://infolux.uni.lu/automatic-speech-recognition-in-luxembourgish-a-very-first-model/) of the University of Luxembourg. 

#### My model has been trained from scratch with my customized dataset [mbarnig/lb-STT_CORPUS](https://huggingface.co/datasets/mbarnig/lb-STT-CORPUS) and the deep-learning-toolkit 🐸 [Coqui-STT](https://github.com/coqui-ai/STT) (version 1.3.0). The model was trained without punctuations with the following alphabet:   
```   
characters="abcdefghijklmnopqrstuvwxyz ßàáâãäçèéêëíîïóôõöùúûü",
punctuations="!'(),-.:;? ",
phonemes=None,
```   
#### A live inference-demo of the ASR system is available in my HuggingFace space ⌨️ 🇱🇺 🔈 [mbarnig/lb-de-fr-en-pt-COQUI-STT](https://huggingface.co/spaces/mbarnig/lb-de-fr-en-pt-COQUI-STT).

#### Click the tab *training metrics* above to view the live Tensorboard of the model training.

![tensorboard](tensorboard.png)