Automatic Speech Recognition
Transformers
Safetensors
Turkish
whisper
asr
speech-recognition
turkish
lora
Eval Results (legacy)
Instructions to use turkmedstt/whisper-large-v3-turkish-general with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use turkmedstt/whisper-large-v3-turkish-general with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("automatic-speech-recognition", model="turkmedstt/whisper-large-v3-turkish-general")# Load model directly from transformers import AutoProcessor, AutoModelForSpeechSeq2Seq processor = AutoProcessor.from_pretrained("turkmedstt/whisper-large-v3-turkish-general") model = AutoModelForSpeechSeq2Seq.from_pretrained("turkmedstt/whisper-large-v3-turkish-general", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Link reproducible 140h data preparation recipe (DATA_PIPELINE.md)
Browse files
README.md
CHANGED
|
@@ -62,6 +62,7 @@ yapısını kullanır; temel deneysel fark M2 eğitimine medikal verinin eklenme
|
|
| 62 |
- Eğitim: 1 epoch
|
| 63 |
- Veri: yaklaşık 140 saat dengeli Türkçe konuşma
|
| 64 |
(Common Voice, ISSAI ve OpenSLR kaynakları)
|
|
|
|
| 65 |
|
| 66 |
## Sonuçlar
|
| 67 |
|
|
|
|
| 62 |
- Eğitim: 1 epoch
|
| 63 |
- Veri: yaklaşık 140 saat dengeli Türkçe konuşma
|
| 64 |
(Common Voice, ISSAI ve OpenSLR kaynakları)
|
| 65 |
+
- Veri hazırlama reçetesi (kaynaklar, lisanslar, temizleme kuralları, sayılar ve yeniden üretim): https://github.com/muhammedkumcu/turkmedstt/blob/main/docs/DATA_PIPELINE.md
|
| 66 |
|
| 67 |
## Sonuçlar
|
| 68 |
|