wav2vec2-large-xlsr-amharic-healthcare

This model is a fine-tuned version of agkphysics/wav2vec2-large-xlsr-53-amharic for Amharic automatic speech recognition (ASR) in the healthcare domain. It is trained to recognize domain-specific vocabulary commonly used in clinical or health-related audio.

🧠 Model Details

• Base model: agkphysics/wav2vec2-large-xlsr-53-amharic
• Language: Amharic (am)
• Domain: Healthcare / Medical
• Training Data: Domain-specific recordings

🛠️ Usage

from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor
import torch
import torchaudio

model = Wav2Vec2ForCTC.from_pretrained("tinkvu/wav2vec2-large-xlsr-amharic-healthcare")
processor = Wav2Vec2Processor.from_pretrained("tinkvu/wav2vec2-large-xlsr-amharic-healthcare")

# Load and preprocess audio
speech_array, sampling_rate = torchaudio.load("your_audio_file.wav")
inputs = processor(speech_array[0], sampling_rate=sampling_rate, return_tensors="pt")

# Perform inference
with torch.no_grad():
    logits = model(**inputs).logits

predicted_ids = torch.argmax(logits, dim=-1)
transcription = processor.batch_decode(predicted_ids)[0]
print(transcription)