Automatic Speech Recognition
Transformers
Safetensors
Turkish
whisper
turkish
asr
speech-recognition
# Load model directly
from transformers import AutoProcessor, AutoModelForSpeechSeq2Seq
processor = AutoProcessor.from_pretrained("RsGoksel/RsGoksel_ITU_Mainframe")
model = AutoModelForSpeechSeq2Seq.from_pretrained("RsGoksel/RsGoksel_ITU_Mainframe", device_map="auto")Quick Links
RsGoksel_ITU_Mainframe
Türkçe ASR — LoRA fine-tune (merged). Çok-domain + robustluk + medikal uyarlama; genel-unseen (ys-0) için optimize.
Kullanım (transformers)
from transformers import pipeline
asr = pipeline("automatic-speech-recognition", model="RsGoksel/RsGoksel_ITU_Mainframe", device=0)
text = asr("audio.wav", generate_kwargs={"language": "tr", "task": "transcribe"})["text"]
Hızlı çıkarım için CTranslate2 sürümü: RsGoksel/RsGoksel_ITU_Mainframe-ct2 (faster-whisper/WhisperX).
Metrik: Türkçe normalizasyon sonrası mean-per-utterance WER.
- Downloads last month
- 35
Model tree for RsGoksel/RsGoksel_ITU_Mainframe
Base model
openai/whisper-large-v3 Finetuned
openai/whisper-large-v3-turbo
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("automatic-speech-recognition", model="RsGoksel/RsGoksel_ITU_Mainframe")