Upload 7 files

.gitattributes

@@ -1,35 +1,35 @@
-*.7z filter=lfs diff=lfs merge=lfs -text
-*.arrow filter=lfs diff=lfs merge=lfs -text
-*.bin filter=lfs diff=lfs merge=lfs -text
-*.bz2 filter=lfs diff=lfs merge=lfs -text
-*.ckpt filter=lfs diff=lfs merge=lfs -text
-*.ftz filter=lfs diff=lfs merge=lfs -text
-*.gz filter=lfs diff=lfs merge=lfs -text
-*.h5 filter=lfs diff=lfs merge=lfs -text
-*.joblib filter=lfs diff=lfs merge=lfs -text
-*.lfs.* filter=lfs diff=lfs merge=lfs -text
-*.mlmodel filter=lfs diff=lfs merge=lfs -text
-*.model filter=lfs diff=lfs merge=lfs -text
-*.msgpack filter=lfs diff=lfs merge=lfs -text
-*.npy filter=lfs diff=lfs merge=lfs -text
-*.npz filter=lfs diff=lfs merge=lfs -text
-*.onnx filter=lfs diff=lfs merge=lfs -text
-*.ot filter=lfs diff=lfs merge=lfs -text
-*.parquet filter=lfs diff=lfs merge=lfs -text
-*.pb filter=lfs diff=lfs merge=lfs -text
-*.pickle filter=lfs diff=lfs merge=lfs -text
-*.pkl filter=lfs diff=lfs merge=lfs -text
-*.pt filter=lfs diff=lfs merge=lfs -text
-*.pth filter=lfs diff=lfs merge=lfs -text
-*.rar filter=lfs diff=lfs merge=lfs -text
-*.safetensors filter=lfs diff=lfs merge=lfs -text
-saved_model/**/* filter=lfs diff=lfs merge=lfs -text
-*.tar.* filter=lfs diff=lfs merge=lfs -text
-*.tar filter=lfs diff=lfs merge=lfs -text
-*.tflite filter=lfs diff=lfs merge=lfs -text
-*.tgz filter=lfs diff=lfs merge=lfs -text
-*.wasm filter=lfs diff=lfs merge=lfs -text
-*.xz filter=lfs diff=lfs merge=lfs -text
-*.zip filter=lfs diff=lfs merge=lfs -text
-*.zst filter=lfs diff=lfs merge=lfs -text
-*tfevents* filter=lfs diff=lfs merge=lfs -text
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text

README.md

@@ -1,13 +1,13 @@
----
-title: Audio Deep Fake
-emoji: 👁
-colorFrom: yellow
-colorTo: purple
-sdk: gradio
-sdk_version: 4.29.0
-app_file: app.py
-pinned: false
-license: mit
----
-
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+---
+title: AudioDeepFake
+emoji: 📊
+colorFrom: purple
+colorTo: yellow
+sdk: gradio
+sdk_version: 4.28.3
+app_file: app.py
+pinned: false
+license: mit
+---
+
+Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

app.py

@@ -0,0 +1,59 @@
+import gradio as gr
+import librosa
+import numpy as np
+import torch
+from torch import Tensor
+import torch.nn as nn
+from model import Model
+
+model_path = 'final_model.pth'
+def load_data(path):
+    X, fs = librosa.load(path)
+    X_pad = pad(X,64600)
+    x_inp = Tensor(X_pad).unsqueeze(0)
+    return x_inp,fs
+
+def pad(x, max_len=64600):
+    x_len = x.shape[0]
+    if x_len >= max_len:
+        return x[:max_len]
+    # need to pad
+    num_repeats = int(max_len / x_len)+1
+    padded_x = np.tile(x, (1, num_repeats))[:, :max_len][0]
+    return padded_x	
+
+device = 'cuda' if torch.cuda.is_available() else 'cpu'
+model = Model(None, device)
+nb_params = sum([param.view(-1).size()[0] for param in model.parameters()])
+model =nn.DataParallel(model).to(device)
+
+model.load_state_dict(torch.load(model_path, map_location=device))
+print("Model loaded : {}".format(model_path))
+
+model.eval()
+prediction_dict = {0: 'Fake', 1: 'Real'}
+def Detection(audio_1):
+    
+    x_inp,fs = load_data(audio_1)
+    print(x_inp.shape)
+    validity_probs = model(x_inp)
+    validity_probs = torch.nn.functional.softmax(validity_probs, dim=1)
+    
+    emotion = torch.argmax(validity_probs).item()
+    print(emotion)
+    validity = prediction_dict[emotion]
+    # validity as a dictionary of class probabilities
+    # validity = {prediction_dict[i]: float(validity_probs[0][i]) for i in range(2)}
+
+    return validity
+
+audio_1 = gr.Audio(type="filepath", label="Audio 1")
+# text_output = gr.Textbox(label="Prediction")
+text_output = gr.Textbox(label="Similarity Score")
+gr.Interface(
+    fn=Detection,
+    inputs=audio_1,
+    outputs=text_output,
+    title="Audio Deepfake Detection",
+    description="Audio Deepfake Detection using finetuned model on for-2seconds dataset.",
+).launch()

final_model.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9b8cbd6c9edd278e22605a9cb58c212405a1364eaa1e145e54aeea1ab06d9ca2
+size 1271630150

model.py

@@ -0,0 +1,597 @@
+import random
+from typing import Union
+
+import numpy as np
+import torch
+import torch.nn as nn
+import torch.nn.functional as F
+from torch import Tensor
+import fairseq
+
+
+___author__ = "Hemlata Tak"
+__email__ = "[email protected]"
+
+############################
+## FOR fine-tuned SSL MODEL
+############################
+
+
+class SSLModel(nn.Module):
+    def __init__(self,device):
+        super(SSLModel, self).__init__()
+        
+        cp_path = 'xlsr2_300m.pt'   # Change the pre-trained XLSR model path. 
+        model, cfg, task = fairseq.checkpoint_utils.load_model_ensemble_and_task([cp_path])
+        self.model = model[0]
+        self.device=device
+        self.out_dim = 1024
+        return
+
+    def extract_feat(self, input_data):
+        
+        # put the model to GPU if it not there
+        if next(self.model.parameters()).device != input_data.device \
+           or next(self.model.parameters()).dtype != input_data.dtype:
+            self.model.to(input_data.device, dtype=input_data.dtype)
+            self.model.train()
+
+        
+        if True:
+            # input should be in shape (batch, length)
+            if input_data.ndim == 3:
+                input_tmp = input_data[:, :, 0]
+            else:
+                input_tmp = input_data
+                
+            # [batch, length, dim]
+            emb = self.model(input_tmp, mask=False, features_only=True)['x']
+        return emb
+
+
+#---------AASIST back-end------------------------#
+''' Jee-weon Jung, Hee-Soo Heo, Hemlata Tak, Hye-jin Shim, Joon Son Chung, Bong-Jin Lee, Ha-Jin Yu and Nicholas Evans. 
+    AASIST: Audio Anti-Spoofing Using Integrated Spectro-Temporal Graph Attention Networks. 
+    In Proc. ICASSP 2022, pp: 6367--6371.'''
+
+
+class GraphAttentionLayer(nn.Module):
+    def __init__(self, in_dim, out_dim, **kwargs):
+        super().__init__()
+
+        # attention map
+        self.att_proj = nn.Linear(in_dim, out_dim)
+        self.att_weight = self._init_new_params(out_dim, 1)
+
+        # project
+        self.proj_with_att = nn.Linear(in_dim, out_dim)
+        self.proj_without_att = nn.Linear(in_dim, out_dim)
+
+        # batch norm
+        self.bn = nn.BatchNorm1d(out_dim)
+
+        # dropout for inputs
+        self.input_drop = nn.Dropout(p=0.2)
+
+        # activate
+        self.act = nn.SELU(inplace=True)
+
+        # temperature
+        self.temp = 1.
+        if "temperature" in kwargs:
+            self.temp = kwargs["temperature"]
+
+    def forward(self, x):
+        '''
+        x   :(#bs, #node, #dim)
+        '''
+        # apply input dropout
+        x = self.input_drop(x)
+
+        # derive attention map
+        att_map = self._derive_att_map(x)
+
+        # projection
+        x = self._project(x, att_map)
+
+        # apply batch norm
+        x = self._apply_BN(x)
+        x = self.act(x)
+        return x
+
+    def _pairwise_mul_nodes(self, x):
+        '''
+        Calculates pairwise multiplication of nodes.
+        - for attention map
+        x           :(#bs, #node, #dim)
+        out_shape   :(#bs, #node, #node, #dim)
+        '''
+
+        nb_nodes = x.size(1)
+        x = x.unsqueeze(2).expand(-1, -1, nb_nodes, -1)
+        x_mirror = x.transpose(1, 2)
+
+        return x * x_mirror
+
+    def _derive_att_map(self, x):
+        '''
+        x           :(#bs, #node, #dim)
+        out_shape   :(#bs, #node, #node, 1)
+        '''
+        att_map = self._pairwise_mul_nodes(x)
+        # size: (#bs, #node, #node, #dim_out)
+        att_map = torch.tanh(self.att_proj(att_map))
+        # size: (#bs, #node, #node, 1)
+        att_map = torch.matmul(att_map, self.att_weight)
+
+        # apply temperature
+        att_map = att_map / self.temp
+
+        att_map = F.softmax(att_map, dim=-2)
+
+        return att_map
+
+    def _project(self, x, att_map):
+        x1 = self.proj_with_att(torch.matmul(att_map.squeeze(-1), x))
+        x2 = self.proj_without_att(x)
+
+        return x1 + x2
+
+    def _apply_BN(self, x):
+        org_size = x.size()
+        x = x.view(-1, org_size[-1])
+        x = self.bn(x)
+        x = x.view(org_size)
+
+        return x
+
+    def _init_new_params(self, *size):
+        out = nn.Parameter(torch.FloatTensor(*size))
+        nn.init.xavier_normal_(out)
+        return out
+
+
+class HtrgGraphAttentionLayer(nn.Module):
+    def __init__(self, in_dim, out_dim, **kwargs):
+        super().__init__()
+
+        self.proj_type1 = nn.Linear(in_dim, in_dim)
+        self.proj_type2 = nn.Linear(in_dim, in_dim)
+
+        # attention map
+        self.att_proj = nn.Linear(in_dim, out_dim)
+        self.att_projM = nn.Linear(in_dim, out_dim)
+
+        self.att_weight11 = self._init_new_params(out_dim, 1)
+        self.att_weight22 = self._init_new_params(out_dim, 1)
+        self.att_weight12 = self._init_new_params(out_dim, 1)
+        self.att_weightM = self._init_new_params(out_dim, 1)
+
+        # project
+        self.proj_with_att = nn.Linear(in_dim, out_dim)
+        self.proj_without_att = nn.Linear(in_dim, out_dim)
+
+        self.proj_with_attM = nn.Linear(in_dim, out_dim)
+        self.proj_without_attM = nn.Linear(in_dim, out_dim)
+
+        # batch norm
+        self.bn = nn.BatchNorm1d(out_dim)
+
+        # dropout for inputs
+        self.input_drop = nn.Dropout(p=0.2)
+
+        # activate
+        self.act = nn.SELU(inplace=True)
+
+        # temperature
+        self.temp = 1.
+        if "temperature" in kwargs:
+            self.temp = kwargs["temperature"]
+
+    def forward(self, x1, x2, master=None):
+        '''
+        x1  :(#bs, #node, #dim)
+        x2  :(#bs, #node, #dim)
+        '''
+        #print('x1',x1.shape)
+        #print('x2',x2.shape)
+        num_type1 = x1.size(1)
+        num_type2 = x2.size(1)
+        #print('num_type1',num_type1)
+        #print('num_type2',num_type2)
+        x1 = self.proj_type1(x1)
+        #print('proj_type1',x1.shape)
+        x2 = self.proj_type2(x2)
+        #print('proj_type2',x2.shape)
+        x = torch.cat([x1, x2], dim=1)
+        #print('Concat x1 and x2',x.shape)
+        
+        if master is None:
+            master = torch.mean(x, dim=1, keepdim=True)
+            #print('master',master.shape)
+        # apply input dropout
+        x = self.input_drop(x)
+
+        # derive attention map
+        att_map = self._derive_att_map(x, num_type1, num_type2)
+        #print('master',master.shape)
+        # directional edge for master node
+        master = self._update_master(x, master)
+        #print('master',master.shape)
+        # projection
+        x = self._project(x, att_map)
+        #print('proj x',x.shape)
+        # apply batch norm
+        x = self._apply_BN(x)
+        x = self.act(x)
+
+        x1 = x.narrow(1, 0, num_type1)
+        #print('x1',x1.shape)
+        x2 = x.narrow(1, num_type1, num_type2)
+        #print('x2',x2.shape)
+        return x1, x2, master
+
+    def _update_master(self, x, master):
+
+        att_map = self._derive_att_map_master(x, master)
+        master = self._project_master(x, master, att_map)
+
+        return master
+
+    def _pairwise_mul_nodes(self, x):
+        '''
+        Calculates pairwise multiplication of nodes.
+        - for attention map
+        x           :(#bs, #node, #dim)
+        out_shape   :(#bs, #node, #node, #dim)
+        '''
+
+        nb_nodes = x.size(1)
+        x = x.unsqueeze(2).expand(-1, -1, nb_nodes, -1)
+        x_mirror = x.transpose(1, 2)
+
+        return x * x_mirror
+
+    def _derive_att_map_master(self, x, master):
+        '''
+        x           :(#bs, #node, #dim)
+        out_shape   :(#bs, #node, #node, 1)
+        '''
+        att_map = x * master
+        att_map = torch.tanh(self.att_projM(att_map))
+
+        att_map = torch.matmul(att_map, self.att_weightM)
+
+        # apply temperature
+        att_map = att_map / self.temp
+
+        att_map = F.softmax(att_map, dim=-2)
+
+        return att_map
+
+    def _derive_att_map(self, x, num_type1, num_type2):
+        '''
+        x           :(#bs, #node, #dim)
+        out_shape   :(#bs, #node, #node, 1)
+        '''
+        att_map = self._pairwise_mul_nodes(x)
+        # size: (#bs, #node, #node, #dim_out)
+        att_map = torch.tanh(self.att_proj(att_map))
+        # size: (#bs, #node, #node, 1)
+
+        att_board = torch.zeros_like(att_map[:, :, :, 0]).unsqueeze(-1)
+
+        att_board[:, :num_type1, :num_type1, :] = torch.matmul(
+            att_map[:, :num_type1, :num_type1, :], self.att_weight11)
+        att_board[:, num_type1:, num_type1:, :] = torch.matmul(
+            att_map[:, num_type1:, num_type1:, :], self.att_weight22)
+        att_board[:, :num_type1, num_type1:, :] = torch.matmul(
+            att_map[:, :num_type1, num_type1:, :], self.att_weight12)
+        att_board[:, num_type1:, :num_type1, :] = torch.matmul(
+            att_map[:, num_type1:, :num_type1, :], self.att_weight12)
+
+        att_map = att_board
+
+        
+
+        # apply temperature
+        att_map = att_map / self.temp
+
+        att_map = F.softmax(att_map, dim=-2)
+
+        return att_map
+
+    def _project(self, x, att_map):
+        x1 = self.proj_with_att(torch.matmul(att_map.squeeze(-1), x))
+        x2 = self.proj_without_att(x)
+
+        return x1 + x2
+
+    def _project_master(self, x, master, att_map):
+
+        x1 = self.proj_with_attM(torch.matmul(
+            att_map.squeeze(-1).unsqueeze(1), x))
+        x2 = self.proj_without_attM(master)
+
+        return x1 + x2
+
+    def _apply_BN(self, x):
+        org_size = x.size()
+        x = x.view(-1, org_size[-1])
+        x = self.bn(x)
+        x = x.view(org_size)
+
+        return x
+
+    def _init_new_params(self, *size):
+        out = nn.Parameter(torch.FloatTensor(*size))
+        nn.init.xavier_normal_(out)
+        return out
+
+
+class GraphPool(nn.Module):
+    def __init__(self, k: float, in_dim: int, p: Union[float, int]):
+        super().__init__()
+        self.k = k
+        self.sigmoid = nn.Sigmoid()
+        self.proj = nn.Linear(in_dim, 1)
+        self.drop = nn.Dropout(p=p) if p > 0 else nn.Identity()
+        self.in_dim = in_dim
+
+    def forward(self, h):
+        Z = self.drop(h)
+        weights = self.proj(Z)
+        scores = self.sigmoid(weights)
+        new_h = self.top_k_graph(scores, h, self.k)
+
+        return new_h
+
+    def top_k_graph(self, scores, h, k):
+        """
+        args
+        =====
+        scores: attention-based weights (#bs, #node, 1)
+        h: graph data (#bs, #node, #dim)
+        k: ratio of remaining nodes, (float)
+        returns
+        =====
+        h: graph pool applied data (#bs, #node', #dim)
+        """
+        _, n_nodes, n_feat = h.size()
+        n_nodes = max(int(n_nodes * k), 1)
+        _, idx = torch.topk(scores, n_nodes, dim=1)
+        idx = idx.expand(-1, -1, n_feat)
+
+        h = h * scores
+        h = torch.gather(h, 1, idx)
+
+        return h
+
+
+
+
+class Residual_block(nn.Module):
+    def __init__(self, nb_filts, first=False):
+        super().__init__()
+        self.first = first
+
+        if not self.first:
+            self.bn1 = nn.BatchNorm2d(num_features=nb_filts[0])
+        self.conv1 = nn.Conv2d(in_channels=nb_filts[0],
+                               out_channels=nb_filts[1],
+                               kernel_size=(2, 3),
+                               padding=(1, 1),
+                               stride=1)
+        self.selu = nn.SELU(inplace=True)
+
+        self.bn2 = nn.BatchNorm2d(num_features=nb_filts[1])
+        self.conv2 = nn.Conv2d(in_channels=nb_filts[1],
+                               out_channels=nb_filts[1],
+                               kernel_size=(2, 3),
+                               padding=(0, 1),
+                               stride=1)
+
+        if nb_filts[0] != nb_filts[1]:
+            self.downsample = True
+            self.conv_downsample = nn.Conv2d(in_channels=nb_filts[0],
+                                             out_channels=nb_filts[1],
+                                             padding=(0, 1),
+                                             kernel_size=(1, 3),
+                                             stride=1)
+
+        else:
+            self.downsample = False
+        
+
+    def forward(self, x):
+        identity = x
+        if not self.first:
+            out = self.bn1(x)
+            out = self.selu(out)
+        else:
+            out = x
+
+        #print('out',out.shape)
+        out = self.conv1(x)
+
+        #print('aft conv1 out',out.shape)
+        out = self.bn2(out)
+        out = self.selu(out)
+        # print('out',out.shape)
+        out = self.conv2(out)
+        #print('conv2 out',out.shape)
+        
+        if self.downsample:
+            identity = self.conv_downsample(identity)
+
+        out += identity
+        #out = self.mp(out)
+        return out
+
+
+class Model(nn.Module):
+    def __init__(self, args,device):
+        super().__init__()
+        self.device = device
+        
+        # AASIST parameters
+        filts = [128, [1, 32], [32, 32], [32, 64], [64, 64]]
+        gat_dims = [64, 32]
+        pool_ratios = [0.5, 0.5, 0.5, 0.5]
+        temperatures =  [2.0, 2.0, 100.0, 100.0]
+
+
+        ####
+        # create network wav2vec 2.0
+        ####
+        self.ssl_model = SSLModel(self.device)
+        self.LL = nn.Linear(self.ssl_model.out_dim, 128)
+
+        self.first_bn = nn.BatchNorm2d(num_features=1)
+        self.first_bn1 = nn.BatchNorm2d(num_features=64)
+        self.drop = nn.Dropout(0.5, inplace=True)
+        self.drop_way = nn.Dropout(0.2, inplace=True)
+        self.selu = nn.SELU(inplace=True)
+
+        # RawNet2 encoder
+        self.encoder = nn.Sequential(
+            nn.Sequential(Residual_block(nb_filts=filts[1], first=True)),
+            nn.Sequential(Residual_block(nb_filts=filts[2])),
+            nn.Sequential(Residual_block(nb_filts=filts[3])),
+            nn.Sequential(Residual_block(nb_filts=filts[4])),
+            nn.Sequential(Residual_block(nb_filts=filts[4])),
+            nn.Sequential(Residual_block(nb_filts=filts[4])))
+
+        self.attention = nn.Sequential(
+            nn.Conv2d(64, 128, kernel_size=(1,1)),
+            nn.SELU(inplace=True),
+            nn.BatchNorm2d(128),
+            nn.Conv2d(128, 64, kernel_size=(1,1)),
+            
+        )
+        # position encoding
+        self.pos_S = nn.Parameter(torch.randn(1, 42, filts[-1][-1]))
+        
+        self.master1 = nn.Parameter(torch.randn(1, 1, gat_dims[0]))
+        self.master2 = nn.Parameter(torch.randn(1, 1, gat_dims[0]))
+        
+        # Graph module
+        self.GAT_layer_S = GraphAttentionLayer(filts[-1][-1],
+                                               gat_dims[0],
+                                               temperature=temperatures[0])
+        self.GAT_layer_T = GraphAttentionLayer(filts[-1][-1],
+                                               gat_dims[0],
+                                               temperature=temperatures[1])
+        # HS-GAL layer 
+        self.HtrgGAT_layer_ST11 = HtrgGraphAttentionLayer(
+            gat_dims[0], gat_dims[1], temperature=temperatures[2])
+        self.HtrgGAT_layer_ST12 = HtrgGraphAttentionLayer(
+            gat_dims[1], gat_dims[1], temperature=temperatures[2])
+        self.HtrgGAT_layer_ST21 = HtrgGraphAttentionLayer(
+            gat_dims[0], gat_dims[1], temperature=temperatures[2])
+        self.HtrgGAT_layer_ST22 = HtrgGraphAttentionLayer(
+            gat_dims[1], gat_dims[1], temperature=temperatures[2])
+
+        # Graph pooling layers
+        self.pool_S = GraphPool(pool_ratios[0], gat_dims[0], 0.3)
+        self.pool_T = GraphPool(pool_ratios[1], gat_dims[0], 0.3)
+        self.pool_hS1 = GraphPool(pool_ratios[2], gat_dims[1], 0.3)
+        self.pool_hT1 = GraphPool(pool_ratios[2], gat_dims[1], 0.3)
+
+        self.pool_hS2 = GraphPool(pool_ratios[2], gat_dims[1], 0.3)
+        self.pool_hT2 = GraphPool(pool_ratios[2], gat_dims[1], 0.3)
+        
+        self.out_layer = nn.Linear(5 * gat_dims[1], 2)
+
+    def forward(self, x):
+        #-------pre-trained Wav2vec model fine tunning ------------------------##
+        x_ssl_feat = self.ssl_model.extract_feat(x.squeeze(-1))
+        x = self.LL(x_ssl_feat) #(bs,frame_number,feat_out_dim)
+        
+        # post-processing on front-end features
+        x = x.transpose(1, 2)   #(bs,feat_out_dim,frame_number)
+        x = x.unsqueeze(dim=1) # add channel 
+        x = F.max_pool2d(x, (3, 3))
+        x = self.first_bn(x)
+        x = self.selu(x)
+
+        # RawNet2-based encoder
+        x = self.encoder(x)
+        x = self.first_bn1(x)
+        x = self.selu(x)
+        
+        w = self.attention(x)
+        
+        #------------SA for spectral feature-------------#
+        w1 = F.softmax(w,dim=-1)
+        m = torch.sum(x * w1, dim=-1)
+        e_S = m.transpose(1, 2) + self.pos_S 
+        
+        # graph module layer
+        gat_S = self.GAT_layer_S(e_S)
+        out_S = self.pool_S(gat_S)  # (#bs, #node, #dim)
+        
+        #------------SA for temporal feature-------------#
+        w2 = F.softmax(w,dim=-2)
+        m1 = torch.sum(x * w2, dim=-2)
+     
+        e_T = m1.transpose(1, 2)
+       
+        # graph module layer
+        gat_T = self.GAT_layer_T(e_T)
+        out_T = self.pool_T(gat_T)
+        
+        # learnable master node
+        master1 = self.master1.expand(x.size(0), -1, -1)
+        master2 = self.master2.expand(x.size(0), -1, -1)
+
+        # inference 1
+        out_T1, out_S1, master1 = self.HtrgGAT_layer_ST11(
+            out_T, out_S, master=self.master1)
+
+        out_S1 = self.pool_hS1(out_S1)
+        out_T1 = self.pool_hT1(out_T1)
+
+        out_T_aug, out_S_aug, master_aug = self.HtrgGAT_layer_ST12(
+            out_T1, out_S1, master=master1)
+        out_T1 = out_T1 + out_T_aug
+        out_S1 = out_S1 + out_S_aug
+        master1 = master1 + master_aug
+
+        # inference 2
+        out_T2, out_S2, master2 = self.HtrgGAT_layer_ST21(
+            out_T, out_S, master=self.master2)
+        out_S2 = self.pool_hS2(out_S2)
+        out_T2 = self.pool_hT2(out_T2)
+
+        out_T_aug, out_S_aug, master_aug = self.HtrgGAT_layer_ST22(
+            out_T2, out_S2, master=master2)
+        out_T2 = out_T2 + out_T_aug
+        out_S2 = out_S2 + out_S_aug
+        master2 = master2 + master_aug
+
+        out_T1 = self.drop_way(out_T1)
+        out_T2 = self.drop_way(out_T2)
+        out_S1 = self.drop_way(out_S1)
+        out_S2 = self.drop_way(out_S2)
+        master1 = self.drop_way(master1)
+        master2 = self.drop_way(master2)
+
+        out_T = torch.max(out_T1, out_T2)
+        out_S = torch.max(out_S1, out_S2)
+        master = torch.max(master1, master2)
+
+        # Readout operation
+        T_max, _ = torch.max(torch.abs(out_T), dim=1)
+        T_avg = torch.mean(out_T, dim=1)
+
+        S_max, _ = torch.max(torch.abs(out_S), dim=1)
+        S_avg = torch.mean(out_S, dim=1)
+        
+        last_hidden = torch.cat(
+            [T_max, T_avg, S_max, S_avg, master.squeeze(1)], dim=1)
+        
+        last_hidden = self.drop(last_hidden)
+        output = self.out_layer(last_hidden)
+        
+        return output

requirements.txt

@@ -0,0 +1,5 @@
+librosa
+numpy
+torch
+torchaudio
+git+https://github.com/KhadgaA/fairseq-a54021305d6b3c4c5959ac9395135f63202db8f1.git

xlsr2_300m.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b08927597f2c9eb2ebd7dcc3ac78ee4b5f6021cbac4b3a6c5a9deec445d80ed9
+size 3808868242