| uv run tools/tokenizer/extend_bpe.py ^ | |
| --base-model checkpoints/bpe.model ^ | |
| --manifests dataset/thaitts_emilia_dataset/forbpe.jsonl ^ | |
| --output-model checkpoints/thai3normalbpe.model ^ | |
| --target-size 36000 ^ | |
| --character-coverage 1.0 ^ | |
| --model-type bpe |
| uv run tools/tokenizer/extend_bpe.py ^ | |
| --base-model checkpoints/bpe.model ^ | |
| --manifests dataset/thaitts_emilia_dataset/forbpe.jsonl ^ | |
| --output-model checkpoints/thai3normalbpe.model ^ | |
| --target-size 36000 ^ | |
| --character-coverage 1.0 ^ | |
| --model-type bpe |