This repository contains the conversion of yuriyvnv/whisper-small-high-mixed-nl to the CTranslate2 model format. This model can be used with faster-whisper.

```python from faster_whisper import WhisperModel

model = WhisperModel("FutureCow/whisper-small-high-mixed-nl-ct2", device="cpu", compute_type="float16")

segments, info = model.transcribe("audio.mp3", language="nl")

for segment in segments: print(f"[{segment.start:.2f}s -> {segment.end:.2f}s] {segment.text}") ```

```bash ct2-transformers-converter
--model yuriyvnv/whisper-small-high-mixed-nl
--output_dir whisper-small-high-mixed-nl-ct2
--copy_files tokenizer.json preprocessor_config.json
--quantization float16 ```