A fine-tuned version of
OpenAI's Whisper Large V3 Turbo optimized for Luxembourgish speech recognition.
This model was fine-tuned on approximately 75 hours of Luxembourgish speech data, curated to be representative of diverse speaking contexts within the available data landscape for the language.
The model is provided in CTranslate2 format for efficient inference with
faster-whisper and compatible libraries.
1from faster_whisper import WhisperModel
2
3model = WhisperModel("ZLSCompLing/WhisperLargeTurboV3_Luxembourgish", device="cuda")
4
5segments, info = model.transcribe("audio.wav", language="lb")
6for segment in segments:
7 print(f"[{segment.start:.2f}s -> {segment.end:.2f}s] {segment.text}")
1import whisperx
2
3model = whisperx.load_model("ZLSCompLing/WhisperLargeTurboV3_Luxembourgish", device="cuda")
4audio = whisperx.load_audio("audio.wav")
5result = model.transcribe(audio, language="lb")
When using this model with WhisperX and PyAnnote for speaker diarization, be aware of potential dependency conflicts between PyTorch, WhisperX, and PyAnnote. Depending on your setup, you may need both modern CUDA libraries and cuDNN 8 libraries installed.
For reference, see the
Sproochmaschinn project, which uses this model in production.
1@misc{zls2025whisperlb,
2 title={Whisper Large V3 Turbo - Luxembourgish},
3 author={Zenter fir d'Lëtzebuerger Sprooch},
4 year={2025},
5 publisher={Hugging Face},
6 url={https://huggingface.co/ZLSCompLing/WhisperLargeTurboV3_Luxembourgish}
7}
This model powers
Sproochmaschinn, a Luxembourgish speech processing platform.
