BriereAssia/wav2vec2_finetuned_darija_80

README.md

@@ -0,0 +1,77 @@
+---
+library_name: transformers
+license: mit
+base_model: facebook/w2v-bert-2.0
+tags:
+- generated_from_trainer
+metrics:
+- wer
+model-index:
+- name: w2v-V2
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# w2v-V2
+
+This model is a fine-tuned version of [facebook/w2v-bert-2.0](https://huggingface.co/facebook/w2v-bert-2.0) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.2363
+- Wer: 0.2199
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 1e-05
+- train_batch_size: 4
+- eval_batch_size: 8
+- seed: 42
+- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: linear
+- training_steps: 5000
+- mixed_precision_training: Native AMP
+
+### Training results
+
+| Training Loss | Epoch  | Step | Validation Loss | Wer    |
+|:-------------:|:------:|:----:|:---------------:|:------:|
+| 3.1759        | 0.1049 | 300  | 0.9128          | 0.8151 |
+| 0.7566        | 0.2099 | 600  | 0.6488          | 0.5731 |
+| 0.6099        | 0.3148 | 900  | 0.5645          | 0.5568 |
+| 0.5685        | 0.4197 | 1200 | 0.4432          | 0.4050 |
+| 0.4799        | 0.5247 | 1500 | 0.4051          | 0.3625 |
+| 0.4532        | 0.6296 | 1800 | 0.4064          | 0.3344 |
+| 0.4197        | 0.7345 | 2100 | 0.3454          | 0.3219 |
+| 0.4358        | 0.8395 | 2400 | 0.3409          | 0.3127 |
+| 0.3889        | 0.9444 | 2700 | 0.3202          | 0.2738 |
+| 0.3637        | 1.0493 | 3000 | 0.2987          | 0.2636 |
+| 0.3367        | 1.1542 | 3300 | 0.2833          | 0.2572 |
+| 0.2684        | 1.2592 | 3600 | 0.2831          | 0.2455 |
+| 0.2909        | 1.3641 | 3900 | 0.2597          | 0.2418 |
+| 0.2995        | 1.4690 | 4200 | 0.2508          | 0.2358 |
+| 0.287         | 1.5740 | 4500 | 0.2423          | 0.2274 |
+| 0.2649        | 1.6789 | 4800 | 0.2363          | 0.2199 |
+
+
+### Framework versions
+
+- Transformers 4.51.1
+- Pytorch 2.5.1+cu121
+- Datasets 3.3.1
+- Tokenizers 0.21.0

config.json

@@ -0,0 +1,81 @@
+{
+  "activation_dropout": 0.0,
+  "adapter_act": "relu",
+  "adapter_kernel_size": 3,
+  "adapter_stride": 2,
+  "add_adapter": true,
+  "apply_spec_augment": false,
+  "architectures": [
+    "Wav2Vec2BertForCTC"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "classifier_proj_size": 768,
+  "codevector_dim": 768,
+  "conformer_conv_dropout": 0.1,
+  "contrastive_logits_temperature": 0.1,
+  "conv_depthwise_kernel_size": 31,
+  "ctc_loss_reduction": "mean",
+  "ctc_zero_infinity": false,
+  "diversity_loss_weight": 0.1,
+  "eos_token_id": 2,
+  "feat_proj_dropout": 0.0,
+  "feat_quantizer_dropout": 0.0,
+  "feature_projection_input_dim": 160,
+  "final_dropout": 0.1,
+  "hidden_act": "swish",
+  "hidden_dropout": 0.0,
+  "hidden_size": 1024,
+  "initializer_range": 0.02,
+  "intermediate_size": 4096,
+  "layer_norm_eps": 1e-05,
+  "layerdrop": 0.0,
+  "left_max_position_embeddings": 64,
+  "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_prob": 0.0,
+  "max_source_positions": 5000,
+  "model_type": "wav2vec2-bert",
+  "num_adapter_layers": 1,
+  "num_attention_heads": 16,
+  "num_codevector_groups": 2,
+  "num_codevectors_per_group": 320,
+  "num_hidden_layers": 24,
+  "num_negatives": 100,
+  "output_hidden_size": 1024,
+  "pad_token_id": 61,
+  "position_embeddings_type": "relative_key",
+  "proj_codevector_dim": 768,
+  "right_max_position_embeddings": 8,
+  "rotary_embedding_base": 10000,
+  "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": "4.51.1",
+  "use_intermediate_ffn_before_adapter": false,
+  "use_weighted_layer_sum": false,
+  "vocab_size": 64,
+  "xvector_output_dim": 512
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0eb21cc38a2c01384777ef718122c77e0d380c20126e89926dca134a84c45214
+size 2423076960

preprocessor_config.json

@@ -0,0 +1,11 @@
+{
+  "feature_extractor_type": "SeamlessM4TFeatureExtractor",
+  "feature_size": 80,
+  "num_mel_bins": 80,
+  "padding_side": "right",
+  "padding_value": 1,
+  "processor_class": "Wav2Vec2BertProcessor",
+  "return_attention_mask": true,
+  "sampling_rate": 16000,
+  "stride": 2
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1073bb0399199f339d2b7577f423183dc307bda4265ce410ca82fb835de86ecc
+size 5240