search

smallflyingpig / universal_adversarial_perturbation_generative_network_for_speaker_recognition

code for paper "Universal Adversarial Perturbations Generative Network for Speaker Recognition"
https://smallflyingpig.github.io/UAPs_for_speaker_recognition/main
MIT License
22 stars 7 forks source link

about targeted attack script #2

Open cdliang11 opened 3 years ago

cdliang11 commented 3 years ago 
python train_generator.py --output_dir ./output/generator/timit_generator_targer100_norm3000_mul01_wlen200_margin0_epoch50_clip001_nolrdecay_fixdmse --speaker_model ./output/SincNet_TIMIT/model_raw.pkl --speaker_cfg ./config/timit_speaker_generator.cfg --noise_scale 1 --norm_factor 3000 --mul 0.1 --norm_clip 0.01 --speaker_factor 1 --wlen 200 --frame_dim 3200 --margin 0 --epoch 50 --data_root ./data/TIMIT/ --dataset timit --num_workers 8 --test_output ./output/generator/timit_generator_target100_norm3000_mul01_wlen200_margin0_epoch50_clip001_nolrdecay_fixdmse_test --no_dist --target 100

hello, this is my targeted attack script, but the result is wrong. Can you share your script? thanks

smallflyingpig commented 3 years ago 
python train_generator.py --output_dir ./output/generator/timit_generator_targer100_norm3000_mul01_wlen200_margin0_epoch50_clip001_nolrdecay_fixdmse --speaker_model ./output/SincNet_TIMIT/model_raw.pkl --speaker_cfg ./config/timit_speaker_generator.cfg --noise_scale 1 --norm_factor 3000 --mul 0.1 --norm_clip 0.01 --speaker_factor 1 --wlen 200 --frame_dim 3200 --margin 0 --epoch 50 --data_root ./data/TIMIT/ --dataset timit --num_workers 8 --test_output ./output/generator/timit_generator_target100_norm3000_mul01_wlen200_margin0_epoch50_clip001_nolrdecay_fixdmse_test --no_dist --target 100

hello, this is my targeted attack script, but the result is wrong. Can you share your script? thanks

Hi, this is your training script? Can you paste the output when something goes wrong?