Views
No views yet
| Metric | Value |
|---|---|
| Validation Word Error Rate (WER) | 0.0558 |
| Test Word Error Rate (WER) | 0.0599 |
| Training Time | < 1 hour on NVIDIA H200 |
| Framework | NVIDIA NeMo |
| Checkpoint Size | 2.34 GB |
| Attribute | Description |
|---|---|
| Base Model | nvidia/parakeet-tdt-0.6b-v3 |
| Dataset | jacktol/ATC-ASR-Dataset |
| Framework | NVIDIA NeMo |
| Epochs | 16 |
| Batch Size | 16 |
| Learning Rate | 1e-4 |
| Optimizer | AdamW (weight decay 1e-3) |
| Scheduler | CosineAnnealing |
| Warmup Steps | 5000 |
| Min LR | 5e-6 |
| Precision | Mixed precision (FP16) |
| Tokenizer | Parakeet default subword tokenizer |
1@misc{nvidia2024parakeet,
2 title={Parakeet-TDT-0.6B-v3},
3 author={NVIDIA},
4 year={2024},
5 publisher={Hugging Face},
6 howpublished={\url{https://huggingface.co/nvidia/parakeet-tdt-0.6b-v3}}
7}
8
9@dataset{jacktol_atc_asr,
10 title={ATC-ASR Dataset},
11 author={Jacktol},
12 year={2023},
13 howpublished={\url{https://huggingface.co/datasets/jacktol/ATC-ASR-Dataset}}
14}