initial commit

36epoch.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:68355225016ff029e259817bb062167c7725f6682cc835c93823ab5ec72383f9
+size 137292312

README.md

@@ -0,0 +1,16 @@
+---
+license: apache-2.0
+---
+
+# IDLS24 TEAM33
+## Attempt at implementing Branch-ECAPA-TDNN
+
+Architecture: [Branch-ECAPA-TDNN](https://www.isca-archive.org/interspeech_2023/yao23_interspeech.html)
+
+Results on Vox1-O, after training on VoxCeleb1-dev
+
+| EER (%) | minDCF|
+|---------|-------|
+|2.654| 0.175 |
+
+eer=3.525, mindcf=0.243

config.yaml

@@ -0,0 +1,203 @@
+config: conf/train_branch_ecapa_mel.yaml
+print_config: false
+log_level: INFO
+drop_last_iter: true
+dry_run: false
+iterator_type: category
+valid_iterator_type: sequence
+output_dir: exp/spk_train_branch_ecapa_mel_raw
+ngpu: 1
+seed: 0
+num_workers: 6
+num_att_plot: 0
+dist_backend: nccl
+dist_init_method: env://
+dist_world_size: null
+dist_rank: null
+local_rank: 0
+dist_master_addr: null
+dist_master_port: null
+dist_launcher: null
+multiprocessing_distributed: false
+unused_parameters: false
+sharded_ddp: false
+cudnn_enabled: true
+cudnn_benchmark: true
+cudnn_deterministic: false
+collect_stats: false
+write_collected_feats: false
+max_epoch: 50
+patience: null
+val_scheduler_criterion:
+- valid
+- loss
+early_stopping_criterion:
+- valid
+- loss
+- min
+best_model_criterion:
+-   - valid
+    - eer
+    - min
+keep_nbest_models: 3
+nbest_averaging_interval: 0
+grad_clip: 9999
+grad_clip_type: 2.0
+grad_noise: false
+accum_grad: 1
+no_forward_run: false
+resume: true
+train_dtype: float32
+use_amp: true
+log_interval: 100
+use_matplotlib: true
+use_tensorboard: true
+create_graph_in_tensorboard: false
+use_wandb: false
+wandb_project: null
+wandb_id: null
+wandb_entity: null
+wandb_name: null
+wandb_model_log_interval: -1
+detect_anomaly: false
+use_lora: false
+save_lora_only: true
+lora_conf: {}
+pretrain_path: null
+init_param: []
+ignore_init_mismatch: false
+freeze_param: []
+num_iters_per_epoch: null
+batch_size: 64
+valid_batch_size: 32
+batch_bins: 1000000
+valid_batch_bins: null
+train_shape_file:
+- exp/spk_stats_16k/train/speech_shape
+valid_shape_file:
+- exp/spk_stats_16k/valid/speech_shape
+batch_type: folded
+valid_batch_type: null
+fold_length:
+- 120000
+sort_in_batch: descending
+shuffle_within_batch: false
+sort_batch: descending
+multiple_iterator: false
+chunk_length: 500
+chunk_shift_ratio: 0.5
+num_cache_chunks: 1024
+chunk_excluded_key_prefixes: []
+chunk_default_fs: null
+train_data_path_and_name_and_type:
+-   - dump/raw/dev_vox1/wav.scp
+    - speech
+    - sound
+-   - dump/raw/dev_vox1/utt2spk
+    - spk_labels
+    - text
+valid_data_path_and_name_and_type:
+-   - dump/raw/test_vox1/trial.scp
+    - speech
+    - sound
+-   - dump/raw/test_vox1/trial2.scp
+    - speech2
+    - sound
+-   - dump/raw/test_vox1/trial_label
+    - spk_labels
+    - text
+allow_variable_data_keys: false
+max_cache_size: 0.0
+max_cache_fd: 32
+allow_multi_rates: false
+valid_max_cache_size: null
+exclude_weight_decay: false
+exclude_weight_decay_conf: {}
+optim: adam
+optim_conf:
+    lr: 1.0e-05
+    weight_decay: 5.0e-05
+    amsgrad: false
+scheduler: reducelronplateau
+scheduler_conf:
+    mode: min
+    factor: 0.5
+    patience: 1
+    min_lr: 1.0e-09
+    threshold: 0.01
+    eps: 1.0e-09
+init: null
+use_preprocessor: true
+input_size: null
+target_duration: 3.0
+spk2utt: dump/raw/dev_vox1/spk2utt
+spk_num: 1211
+sample_rate: 16000
+num_eval: 10
+rir_scp: ''
+model_conf:
+    extract_feats_in_collect_stats: false
+frontend: melspec_torch
+frontend_conf:
+    preemp: true
+    n_fft: 512
+    log: true
+    win_length: 400
+    hop_length: 160
+    n_mels: 80
+    normalize: mn
+specaug: null
+specaug_conf: {}
+normalize: null
+normalize_conf: {}
+encoder: branch_ecapa_tdnn
+encoder_conf:
+    model_scale: 8
+    ndim: 1024
+    output_size: 1536
+    num_heads: 2
+    dropout_rate: 0.4
+    merge_method: concat
+pooling: chn_attn_stat
+pooling_conf: {}
+projector: rawnet3
+projector_conf:
+    output_size: 192
+preprocessor: spk
+preprocessor_conf:
+    target_duration: 3.0
+    sample_rate: 16000
+    num_eval: 5
+    noise_apply_prob: 0.5
+    noise_info:
+    -   - 1.0
+        - dump/raw/musan_speech.scp
+        -   - 4
+            - 7
+        -   - 13
+            - 20
+    -   - 1.0
+        - dump/raw/musan_noise.scp
+        -   - 1
+            - 1
+        -   - 0
+            - 15
+    -   - 1.0
+        - dump/raw/musan_music.scp
+        -   - 1
+            - 1
+        -   - 5
+            - 15
+    rir_apply_prob: 0.5
+    rir_scp: dump/raw/rirs.scp
+loss: aamsoftmax_sc_topk
+loss_conf:
+    margin: 0.3
+    scale: 30
+    K: 3
+    mp: 0.06
+    k_top: 5
+required:
+- output_dir
+version: '202402'
+distributed: false