feat: update model config

added_tokens.json

@@ -1 +0,0 @@
-{"<s>": 37, "</s>": 38}

config.json

@@ -1,107 +0,0 @@
-{
-  "_name_or_path": "facebook/wav2vec2-xls-r-300m",
-  "activation_dropout": 0.0,
-  "adapter_kernel_size": 3,
-  "adapter_stride": 2,
-  "add_adapter": false,
-  "apply_spec_augment": true,
-  "architectures": [
-    "Wav2Vec2ForCTC"
-  ],
-  "attention_dropout": 0.0,
-  "bos_token_id": 1,
-  "classifier_proj_size": 256,
-  "codevector_dim": 768,
-  "contrastive_logits_temperature": 0.1,
-  "conv_bias": true,
-  "conv_dim": [
-    512,
-    512,
-    512,
-    512,
-    512,
-    512,
-    512
-  ],
-  "conv_kernel": [
-    10,
-    3,
-    3,
-    3,
-    3,
-    2,
-    2
-  ],
-  "conv_stride": [
-    5,
-    2,
-    2,
-    2,
-    2,
-    2,
-    2
-  ],
-  "ctc_loss_reduction": "mean",
-  "ctc_zero_infinity": false,
-  "diversity_loss_weight": 0.1,
-  "do_stable_layer_norm": true,
-  "eos_token_id": 2,
-  "feat_extract_activation": "gelu",
-  "feat_extract_dropout": 0.0,
-  "feat_extract_norm": "layer",
-  "feat_proj_dropout": 0.0,
-  "feat_quantizer_dropout": 0.0,
-  "final_dropout": 0.0,
-  "hidden_act": "gelu",
-  "hidden_dropout": 0.0,
-  "hidden_size": 1024,
-  "initializer_range": 0.02,
-  "intermediate_size": 4096,
-  "layer_norm_eps": 1e-05,
-  "layerdrop": 0.0,
-  "mask_feature_length": 64,
-  "mask_feature_min_masks": 0,
-  "mask_feature_prob": 0.25,
-  "mask_time_length": 10,
-  "mask_time_min_masks": 2,
-  "mask_time_prob": 0.75,
-  "model_type": "wav2vec2",
-  "num_adapter_layers": 3,
-  "num_attention_heads": 16,
-  "num_codevector_groups": 2,
-  "num_codevectors_per_group": 320,
-  "num_conv_pos_embedding_groups": 16,
-  "num_conv_pos_embeddings": 128,
-  "num_feat_extract_layers": 7,
-  "num_hidden_layers": 24,
-  "num_negatives": 100,
-  "output_hidden_size": 1024,
-  "pad_token_id": 36,
-  "proj_codevector_dim": 768,
-  "tdnn_dilation": [
-    1,
-    2,
-    3,
-    1,
-    1
-  ],
-  "tdnn_dim": [
-    512,
-    512,
-    512,
-    512,
-    1500
-  ],
-  "tdnn_kernel": [
-    5,
-    3,
-    3,
-    1,
-    1
-  ],
-  "torch_dtype": "float32",
-  "transformers_version": "4.16.0.dev0",
-  "use_weighted_layer_sum": false,
-  "vocab_size": 39,
-  "xvector_output_dim": 512
-}

preprocessor_config.json

@@ -1,9 +0,0 @@
-{
-  "do_normalize": true,
-  "feature_extractor_type": "Wav2Vec2FeatureExtractor",
-  "feature_size": 1,
-  "padding_side": "right",
-  "padding_value": 0,
-  "return_attention_mask": true,
-  "sampling_rate": 16000
-}

pytorch_model.bin

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:122d5480924adf4705e1f37e28c0c798df35e76a41767b73474d3d5df8c1062f
-size 1262083569

run.sh

@@ -1,36 +1,38 @@
 #!/bin/sh
 
 export CUDA_VISIBLE_DEVICES=1,2
+
 python src/run_speech_recognition_ctc_bnb.py \
        --dataset_name="mozilla-foundation/common_voice_7_0" \
-       --model_name_or_path="facebook/wav2vec2-xls-r-300m" \
+       --model_name_or_path="facebook/wav2vec2-xls-r-1b" \
        --dataset_config_name="et" \
        --output_dir="./" \
        --overwrite_output_dir \
-       --num_train_epochs="50" \
-       --per_device_train_batch_size="16" \
-       --per_device_eval_batch_size="16" \
-       --gradient_accumulation_steps="2" \
-       --learning_rate="7.5e-5" \
-       --save_total_limit="3" \
-       --warmup_steps="2000" \
-       --evaluation_strategy="steps" \
-       --text_column_name="sentence" \
-       --length_column_name="input_length" \
-       --save_steps="500" \
-       --eval_steps="500" \
-       --logging_steps="100" \
-       --layerdrop="0.0" \
+       --num_train_epochs=100 \
+       --per_device_train_batch_size=32 \
+       --per_device_eval_batch_size=32 \
+       --gradient_accumulation_steps=2 \
+       --learning_rate=3e-4 \
+       --save_total_limit=3 \
+       --warmup_steps=500 \
+       --evaluation_strategy=steps \
+       --text_column_name=sentence \
+       --length_column_name=input_length \
+       --save_steps=500 \
+       --eval_steps=500 \
+       --logging_steps=100 \
+       --layerdrop=0.0 \
        --freeze_feature_encoder \
-       --feat_proj_dropout="0.0" \
-       --mask_time_prob="0.75" \
-       --mask_time_length="10" \
-       --mask_feature_prob="0.25" \
-       --mask_feature_length="64" \
+       --feat_proj_dropout=0.1 \
        --chars_to_ignore , ? . ! \- \; \: \" “ % ‘ ” � — ’ … – \
        --gradient_checkpointing \
-       --use_auth_token \
+       --lr_scheduler_type=cosine \
        --fp16 \
        --group_by_length \
+       --mask_time_prob=0.1 \
+       --mask_time_length=10 \
+       --report_to=wandb \
+       --run_name="cosine+drop_proj+low_specaugment-1b" \
        --do_train --do_eval \
-       --push_to_hub
+       --use_auth_token \
+       --push_to_hub

special_tokens_map.json

@@ -1 +0,0 @@
-{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "[UNK]", "pad_token": "[PAD]", "additional_special_tokens": [{"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}]}

tokenizer_config.json

@@ -1 +0,0 @@
-{"unk_token": "[UNK]", "bos_token": "<s>", "eos_token": "</s>", "pad_token": "[PAD]", "do_lower_case": false, "word_delimiter_token": "|", "special_tokens_map_file": null, "tokenizer_file": null, "name_or_path": "./", "tokenizer_class": "Wav2Vec2CTCTokenizer"}

training_args.bin

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:f3e82d9fe00fea24d2f00c4b477dbd5b91e33cea6ba3cf8903f298d8f42fefbe
-size 2991

vocab.json

@@ -1 +0,0 @@
-{"a": 1, "b": 2, "c": 3, "d": 4, "e": 5, "f": 6, "g": 7, "h": 8, "i": 9, "j": 10, "k": 11, "l": 12, "m": 13, "n": 14, "o": 15, "p": 16, "q": 17, "r": 18, "s": 19, "t": 20, "u": 21, "v": 22, "w": 23, "x": 24, "y": 25, "z": 26, "ä": 27, "õ": 28, "ö": 29, "ü": 30, "š": 31, "ž": 32, "̇": 33, "„": 34, "|": 0, "[UNK]": 35, "[PAD]": 36}