Update model

README.md

@@ -61,6 +61,7 @@ sharded_ddp: false
 cudnn_enabled: true
 cudnn_benchmark: false
 cudnn_deterministic: false
+use_tf32: false
 collect_stats: false
 write_collected_feats: false
 max_epoch: 120
@@ -92,7 +93,7 @@ no_forward_run: false
 resume: true
 train_dtype: float32
 use_amp: false
-log_interval: 500
+log_interval: 5
 use_matplotlib: true
 use_tensorboard: true
 create_graph_in_tensorboard: false
@@ -117,9 +118,9 @@ valid_batch_size: null
 batch_bins: 1000000
 valid_batch_bins: null
 train_shape_file:
-- exp/codec_stats_raw/train/audio_shape
+- exp/stats_16k/train/audio_shape
 valid_shape_file:
-- exp/codec_stats_raw/valid/audio_shape
+- exp/stats_16k/valid/audio_shape
 batch_type: unsorted
 valid_batch_type: null
 fold_length:
@@ -141,6 +142,7 @@ valid_data_path_and_name_and_type:
 -   - dump_16k/raw/dev-clean/wav.scp
     - audio
     - sound
+multi_task_dataset: false
 allow_variable_data_keys: false
 max_cache_size: 0.0
 max_cache_fd: 32

exp/codec_encodec_ss4_16k/120epoch.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a7e1d2abc857c8da7a8207fbff1b5bdcfa2e7b4d279f3d65c699b31cb539303b
+size 215606502

exp/codec_encodec_ss4_16k/config.yaml

@@ -24,6 +24,7 @@ sharded_ddp: false
 cudnn_enabled: true
 cudnn_benchmark: false
 cudnn_deterministic: false
+use_tf32: false
 collect_stats: false
 write_collected_feats: false
 max_epoch: 120
@@ -55,7 +56,7 @@ no_forward_run: false
 resume: true
 train_dtype: float32
 use_amp: false
-log_interval: 500
+log_interval: 5
 use_matplotlib: true
 use_tensorboard: true
 create_graph_in_tensorboard: false
@@ -80,9 +81,9 @@ valid_batch_size: null
 batch_bins: 1000000
 valid_batch_bins: null
 train_shape_file:
-- exp/codec_stats_raw/train/audio_shape
+- exp/stats_16k/train/audio_shape
 valid_shape_file:
-- exp/codec_stats_raw/valid/audio_shape
+- exp/stats_16k/valid/audio_shape
 batch_type: unsorted
 valid_batch_type: null
 fold_length:
@@ -104,6 +105,7 @@ valid_data_path_and_name_and_type:
 -   - dump_16k/raw/dev-clean/wav.scp
     - audio
     - sound
+multi_task_dataset: false
 allow_variable_data_keys: false
 max_cache_size: 0.0
 max_cache_fd: 32

meta.yaml

@@ -1,8 +1,8 @@
 espnet: '202402'
 files:
-  model_file: exp/codec_encodec_ss4_16k/81epoch.pth
+  model_file: exp/codec_encodec_ss4_16k/120epoch.pth
 python: 3.10.14 (main, May  6 2024, 19:42:50) [GCC 11.2.0]
-timestamp: 1718729736.568983
+timestamp: 1718989322.954464
 torch: 2.0.1
 yaml_files:
   train_config: exp/codec_encodec_ss4_16k/config.yaml