metadata

language:
  - pap
license: apache-2.0
tags:
  - whisper
  - automatic-speech-recognition
  - papiamento
  - speech-to-text
  - medical
  - healthcare
  - clinical
base_model: sonnygeorge/whisper-tiny-pap
datasets:
  - medical-papiamento-corpus
widget:
  - example_title: Medical Papiamento Sample
    src: https://example.com/medical_sample.wav

Whisper Tiny Papiamento - Medical Domain Adaptation

This model is a medical domain fine-tuned version of sonnygeorge/whisper-tiny-pap specialized for healthcare and clinical Papiamento speech recognition.

Model Description

Base model: sonnygeorge/whisper-tiny-pap (Papiamento Whisper model by Sonny George)
Domain: Medical/Healthcare Papiamento
Language: Papiamento (pap)
Specialization: Clinical terminology, medical consultations, healthcare vocabulary
Training: Fine-tuned on medical Papiamento audio data

Model Performance

This model builds upon Sonny George's excellent Papiamento Whisper foundation and adds:

✅ Enhanced medical terminology recognition
✅ Clinical context understanding
✅ Healthcare vocabulary optimization
✅ Single speaker adaptation for consistent medical speech patterns

Intended Uses

Medical consultation transcription in Papiamento
Clinical note generation from Papiamento audio
Healthcare documentation automation
Medical terminology recognition in Papiamento

Usage

from transformers import WhisperProcessor, WhisperForConditionalGeneration
import librosa

# Load model and processor
processor = WhisperProcessor.from_pretrained("your-username/whisper-tiny-pap-medical")
model = WhisperForConditionalGeneration.from_pretrained("your-username/whisper-tiny-pap-medical")

# Load medical audio
audio, sr = librosa.load("medical_consultation.m4a", sr=16000)

# Process and transcribe
inputs = processor(audio, sampling_rate=16000, return_tensors="pt")
predicted_ids = model.generate(inputs["input_features"])
transcription = processor.batch_decode(predicted_ids, skip_special_tokens=True)[0]

print(transcription)  # Medical Papiamento transcription