Views
No views yet
Phonepadith/laos-speech-dataset, a curated dataset containing Lao speech samples and transcriptions.| Property | Description |
|---|---|
| Base model | openai/whisper-v3-large |
| Fine-tuned by | @Phonepadith |
| Language | Lao (lo) |
| Task | Automatic Speech Recognition (ASR) |
| Framework | 🤗 Transformers, PyTorch |
| Dataset | Phonepadith/laos-speech-dataset |
| Sampling rate | 16 kHz |
| License | MIT (same as base model unless otherwise stated) |
1from transformers import WhisperProcessor, WhisperForConditionalGeneration
2import torch
3import torchaudio
4
5# Load model and processor
6model_id = "Phonepadith/whisper-3-large-lao-finetuned-v1"
7processor = WhisperProcessor.from_pretrained(model_id)
8model = WhisperForConditionalGeneration.from_pretrained(model_id)
9
10# Load an audio file (16kHz mono)
11speech_array, sampling_rate = torchaudio.load("example.wav")
12speech_array = torchaudio.functional.resample(speech_array, sampling_rate, 16000)
13
14# Preprocess and generate transcription
15input_features = processor(
16 speech_array.squeeze().numpy(),
17 sampling_rate=16000,
18 return_tensors="pt"
19).input_features
20
21predicted_ids = model.generate(input_features)
22transcription = processor.batch_decode(predicted_ids, skip_special_tokens=True)[0]
23
24print("Transcription:", transcription)| Step | Training Loss | Validation Loss |
|---|---|---|
| 500 | 0.124200 | 0.109733 |
| 1000 | 0.051000 | 0.055065 |
| 1500 | 0.034800 | 0.040616 |
| 2000 | 0.023100 | 0.033179 |
| 2500 | 0.016200 | 0.027788 |
| 3000 | 0.007700 | 0.026611 |
| 3500 | 0.007100 | 0.023043 |
| 4000 | 0.003000 | 0.021656 |
| 4500 | 0.002200 | 0.020975 |
| 5000 | 0.001100 | 0.020395 |
1@misc{phonepadith2025whisperlao,
2 title = {Whisper Large Fine-tuned for Lao ASR},
3 author = {Phonepadith Phoummavong},
4 year = {2025},
5 howpublished = {\url{https://huggingface.co/Phonepadith/whisper-3-large-lao-finetuned-v1}},
6}