Download training_config.yaml from RootDeveloperDS/Visar_OpenWakeWord_Model: direct link, hf CLI and curl.
- Browser
- Download file 759 Bytes
-
https://huggingface.co/RootDeveloperDS/Visar_OpenWakeWord_Model/resolve/main/training_config.yaml
- Command line
-
hf download hf://RootDeveloperDS/Visar_OpenWakeWord_Model/training_config.yaml
-
curl -L -o training_config.yaml https://huggingface.co/RootDeveloperDS/Visar_OpenWakeWord_Model/resolve/main/training_config.yaml
759 Bytes
| augmentation_batch_size: 16 | |
| augmentation_rounds: 1 | |
| background_paths: | |
| - ./audioset_16k | |
| - ./fma | |
| background_paths_duplication_rate: | |
| - 1 | |
| batch_n_per_class: | |
| ACAV100M_sample: 1024 | |
| adversarial_negative: 50 | |
| positive: 50 | |
| custom_negative_phrases: [] | |
| false_positive_validation_data_path: validation_set_features.npy | |
| feature_data_files: | |
| ACAV100M_sample: openwakeword_features_ACAV100M_2000_hrs_16bit.npy | |
| layer_size: 32 | |
| max_negative_weight: 3000 | |
| model_name: visar_edge | |
| model_type: dnn | |
| n_samples: 5000 | |
| n_samples_val: 500 | |
| output_dir: ./my_custom_model | |
| piper_sample_generator_path: ./piper-sample-generator | |
| rir_paths: | |
| - ./mit_rirs | |
| steps: 35000 | |
| target_accuracy: 0.5 | |
| target_false_positives_per_hour: 0.2 | |
| target_phrase: | |
| - vee_saar | |
| target_recall: 0.25 | |
| tts_batch_size: 50 | |