Downloads · 30 days
939
70% of all-time downloads
yuriyvnv/whisper-large-v3-slovenian
whisper-large-v3-slovenian is a automatic speech recognition model from yuriyvnv. Use it when you need speech turned into text. It is set up for transformers. The card lists the license as apache-2.0.
Fine-tuned OpenAI Whisper-large-v3 for Slovenian automatic speech recognition, augmented with TTS-generated synthetic data.
Downloads · 30 days
939
70% of all-time downloads
All-time downloads
1.3K
Public
Parameters
1.5B
6.2 GB on disk
Likes
1
Public
Click a slice to open those files.
.safetensors6.2 GB · 100%
From the Hugging Face model README
Fine-tuned OpenAI Whisper-large-v3 for Slovenian automatic speech recognition, augmented with TTS-generated synthetic data.
This model is part of the paper Synthetic speech augmentation for low-resource Estonian and Slovenian ASR: Comparing Parakeet-TDT and Whisper, published in Machine Learning with Applications 26 (2026) 101043.
Note: Fine-tuning Whisper on CommonVoice data degrades FLEURS performance for Slovenian. Zero-shot Whisper (37.02% raw / 19.18% normalized) outperforms all fine-tuned variants on FLEURS. This is because Whisper's pre-trained model handles FLEURS-style punctuation and casing, which fine-tuning on CV-style data erodes. On the in-domain CV17 test set, this model provides substantial improvements over both zero-shot and CV-only fine-tuning.
openai/whisper-large-v3| Test Set | WER | CER |
|---|---|---|
| CommonVoice 17 Test | 15.65 | 3.88 |
| CommonVoice 17 Val | 15.08 | 3.60 |
| FLEURS Test | 40.46 | 11.98 |
| Test Set | WER | CER |
|---|---|---|
| CommonVoice 17 Test | 13.24 | 3.48 |
| CommonVoice 17 Val | 12.98 | 3.24 |
| FLEURS Test | 24.52 | 8.32 |
| Comparison | CV17 Test (WER) | FLEURS Test (WER) |
|---|---|---|
| vs. Zero-shot | -5.55 pp | +3.44 pp* |
| vs. CV-only fine-tuning | -3.66 pp | -6.33 pp |
*FLEURS degrades vs. zero-shot due to domain mismatch (see note above), but improves substantially vs. CV-only fine-tuning.
All CV17 improvements are statistically significant (paired bootstrap, p < 0.001, n = 100,000).
from transformers import WhisperForConditionalGeneration, WhisperProcessor
import torch
# Load model and processor
model = WhisperForConditionalGeneration.from_pretrained("yuriyvnv/whisper-large-v3-slovenian")
processor = WhisperProcessor.from_pretrained("yuriyvnv/whisper-large-v3-slovenian")
# Load audio (16kHz mono)
import librosa
audio, sr = librosa.load("audio.wav", sr=16000)
# Transcribe
input_features = processor(audio, sampling_rate=16000, return_tensors="pt").input_features
with torch.no_grad():
predicted_ids = model.generate(input_features, language="sl", task="transcribe")
transcription = processor.batch_decode(predicted_ids, skip_special_tokens=True)[0]
print(transcription)
The synthetic training data was generated using a three-stage pipeline:
Dataset: yuriyvnv/synthetic_asr_et_sl
@article{PEREZHOHIN2026101043,
title = {Synthetic speech augmentation for low-resource Estonian and Slovenian ASR: Comparing Parakeet-TDT and Whisper},
journal = {Machine Learning with Applications},
volume = {26},
pages = {101043},
year = {2026},
issn = {2666-8270},
doi = {10.1016/j.mlwa.2026.101043},
url = {https://www.sciencedirect.com/science/article/pii/S2666827026002082},
author = {Yuriy Perezhohin and Mauro Castelli and Aleš Popovič},
}