wav2vec2phoneme

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "facebook/wav2vec2-xls-r-300m",
+  "_name_or_path": "facebook/wav2vec2-xlsr-53-espeak-cv-ft",
   "activation_dropout": 0.0,
   "adapter_kernel_size": 3,
   "adapter_stride": 2,
@@ -60,12 +60,20 @@
   "intermediate_size": 4096,
   "layer_norm_eps": 1e-05,
   "layerdrop": 0.0,
+  "mask_channel_length": 10,
+  "mask_channel_min_space": 1,
+  "mask_channel_other": 0.0,
+  "mask_channel_prob": 0.0,
+  "mask_channel_selection": "static",
   "mask_feature_length": 10,
   "mask_feature_min_masks": 0,
   "mask_feature_prob": 0.0,
   "mask_time_length": 10,
   "mask_time_min_masks": 2,
+  "mask_time_min_space": 1,
+  "mask_time_other": 0.0,
   "mask_time_prob": 0.05,
+  "mask_time_selection": "static",
   "model_type": "wav2vec2",
   "num_adapter_layers": 3,
   "num_attention_heads": 16,
@@ -77,7 +85,7 @@
   "num_hidden_layers": 24,
   "num_negatives": 100,
   "output_hidden_size": 1024,
-  "pad_token_id": 40,
+  "pad_token_id": 0,
   "proj_codevector_dim": 768,
   "tdnn_dilation": [
     1,
@@ -103,6 +111,6 @@
   "torch_dtype": "float32",
   "transformers_version": "4.25.1",
   "use_weighted_layer_sum": false,
-  "vocab_size": 43,
+  "vocab_size": 392,
   "xvector_output_dim": 512
 }

preprocessor_config.json

@@ -3,7 +3,7 @@
   "feature_extractor_type": "Wav2Vec2FeatureExtractor",
   "feature_size": 1,
   "padding_side": "right",
-  "padding_value": 0.0,
+  "padding_value": 0,
   "return_attention_mask": true,
   "sampling_rate": 16000
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bea138120341524c31a48a317b1c3f14d95ebff52ecaca81985931b040d593f0
-size 1262074993
+oid sha256:662e2b92f64f89af6da953d1ab6988f87b585a4a8bd868e5df0f52de4379ccd6
+size 1263505905

runs/Jan08_06-16-12_899a1481aea2/1673159019.6706855/events.out.tfevents.1673159019.899a1481aea2.103.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9b81f163ee6ae835dc38328277f7b339cff0756b927e1d7002af7b00cacf8406
+size 5531

runs/Jan08_06-16-12_899a1481aea2/events.out.tfevents.1673159019.899a1481aea2.103.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:134fcffe6c63c2886ac518bed5b66eba3b54d225c2520c0f4a1561a268a1be13
+size 5760

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9cc3a50c7d3fde6950d0958a54bd4d60b864103f144b186b7d018cb1bf380b3a
+oid sha256:f9e56e341c1acc4a0c79909f815bdc438181fc935f6add6bf3f781bc6e004df2
 size 3375