thantzinphyo/whisper-tiny-myanmar-shopvoice

Whisper Tiny Myanmar ShopVoice

This model is a further fine-tuned version of thantzinphyo/whisper-tiny-myanmar-shopvoice for myanmar shopvoice speech recognition.

The ShopVoice model was trained for Myanmar shop speech recognition. This updated version continues training with additional shop-domain speech data to improve recognition of natural Myanmar shop conversations and ordering expressions.

Base Model

• Initial Myanmar Whisper model: chuuhtetnaing/whisper-tiny-myanmar
• Continued ShopVoice model: thantzinphyo/whisper-tiny-myanmar-shopvoice
• Dataset: thantzinphyo/myanmar-shopvoice

Training Results

Step	Training Loss	Validation Loss	WER
100	0.413740	0.108490	65.882353
200	0.090619	0.063400	44.705882
300	0.049589	0.059023	40.000000
400	0.027028	0.059179	38.235294

Usage

from transformers import pipeline

pipe = pipeline(
    "automatic-speech-recognition",
    model="thantzinphyo/whisper-tiny-myanmar-shopvoice",
)

output = pipe(
    "your_audio.wav",
    generate_kwargs={"language": "myanmar", "task": "transcribe"},
)
print(output["text"]) # ကြက်သားဟင်းတစ်ပွဲ ဆိတ်သားဆီပြန်ခြောက်ပွဲ ချဉ်ပေါင်ရွက်တစ်စီး

Intended Use

This model is intended for learning and research on Myanmar speech recognition in real-world shop environments, focusing on natural shop conversations and transactional speech.

Notes

This is a domain fine-tuned Whisper Tiny model. Accuracy can vary depending on recording quality, accent, noise, and whether the spoken words are close to the training domain.