Add OpenVINO model

README.md

@@ -0,0 +1,31 @@
+---
+language:
+- en
+tags:
+- openvino
+---
+
+# anton-l/wav2vec2-base-superb-sd
+
+This is the [anton-l/wav2vec2-base-superb-sd](https://huggingface.co/anton-l/wav2vec2-base-superb-sd) model, converted
+to OpenVINO. An example of how to do inference on this model:
+
+```python
+from transformers import AutoFeatureExtractor
+from optimum.intel.openvino import OVModelForAudioFrameClassification
+from datasets import load_dataset
+import torch
+
+dataset = load_dataset("hf-internal-testing/librispeech_asr_demo", "clean", split="validation")
+dataset = dataset.sort("id")
+sampling_rate = dataset.features["audio"].sampling_rate
+
+feature_extractor = AutoFeatureExtractor.from_pretrained("helenai/anton-l-wav2vec2-base-superb-sd-ov")
+model =  OVModelForAudioFrameClassification.from_pretrained("helenai/anton-l-wav2vec2-base-superb-sd-ov")
+inputs = feature_extractor(dataset[0]["audio"]["array"], return_tensors="np", sampling_rate=sampling_rate)
+logits = model(**inputs).logits
+
+probabilities = torch.sigmoid(torch.as_tensor(logits)[0])
+labels = (probabilities > 0.5).long()
+print(labels[0].tolist())
+```

config.json

@@ -0,0 +1,125 @@
+{
+  "_name_or_path": "anton-l/wav2vec2-base-superb-sd",
+  "activation_dropout": 0.0,
+  "adapter_kernel_size": 3,
+  "adapter_stride": 2,
+  "add_adapter": false,
+  "apply_spec_augment": true,
+  "architectures": [
+    "Wav2Vec2ForAudioFrameClassification"
+  ],
+  "attention_dropout": 0.1,
+  "bos_token_id": 1,
+  "classifier_proj_size": 256,
+  "codevector_dim": 256,
+  "contrastive_logits_temperature": 0.1,
+  "conv_bias": false,
+  "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": "sum",
+  "ctc_zero_infinity": false,
+  "diversity_loss_weight": 0.1,
+  "do_stable_layer_norm": false,
+  "eos_token_id": 2,
+  "feat_extract_activation": "gelu",
+  "feat_extract_norm": "group",
+  "feat_proj_dropout": 0.1,
+  "feat_quantizer_dropout": 0.0,
+  "final_dropout": 0.0,
+  "freeze_feat_extract_train": true,
+  "hidden_act": "gelu",
+  "hidden_dropout": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "speaker_0",
+    "1": "speaker_1"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "speaker_0": 0,
+    "speaker_1": 1
+  },
+  "layer_norm_eps": 1e-05,
+  "layerdrop": 0.05,
+  "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",
+  "no_mask_channel_overlap": false,
+  "no_mask_time_overlap": false,
+  "num_adapter_layers": 3,
+  "num_attention_heads": 12,
+  "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": 12,
+  "num_negatives": 100,
+  "output_hidden_size": 768,
+  "pad_token_id": 0,
+  "proj_codevector_dim": 256,
+  "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": "4.28.1",
+  "use_weighted_layer_sum": true,
+  "vocab_size": 32,
+  "xvector_output_dim": 512
+}

openvino_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:561fc066c6afba3c5fa987b2a022ba1a4721856e2a179569cb454a63a528b0f6
+size 377493804