base/base-encoder_fp32/model.txt

/audioEncoder/conv1/Conv:Conv*input:mel(0,80,0);onnx_3A__3A_Conv_5F_1218_nchw.bin(float32:512,80,3,1);onnx_3A__3A_Conv_5F_1219.bin(float32:512)*output:_2F_audioEncoder_2F_conv1_2F_Conv_5F_output_5F_0(0,512,0)*dilations:1;group:1;kernel_shape:3;pads:1,1;strides:1
/audioEncoder/Div:Div*input:_2F_audioEncoder_2F_conv1_2F_Conv_5F_output_5F_0(0,512,0);_2F_audioEncoder_2F_Constant_5F_output_5F_0.bin(float32:)*output:_2F_audioEncoder_2F_Div_5F_output_5F_0(0,512,0)
/audioEncoder/Erf:Erf*input:_2F_audioEncoder_2F_Div_5F_output_5F_0(0,512,0)*output:_2F_audioEncoder_2F_Erf_5F_output_5F_0(0,512,0)
/audioEncoder/Add:Add*input:_2F_audioEncoder_2F_Erf_5F_output_5F_0(0,512,0);_2F_audioEncoder_2F_Constant_5F_1_5F_output_5F_0.bin(float32:)*output:_2F_audioEncoder_2F_Add_5F_output_5F_0(0,512,0)
/audioEncoder/Mul:Mul*input:_2F_audioEncoder_2F_conv1_2F_Conv_5F_output_5F_0(0,512,0);_2F_audioEncoder_2F_Add_5F_output_5F_0(0,512,0)*output:_2F_audioEncoder_2F_Mul_5F_output_5F_0(0,512,0)
/audioEncoder/Mul_1:Mul*input:_2F_audioEncoder_2F_Mul_5F_output_5F_0(0,512,0);_2F_audioEncoder_2F_Constant_5F_2_5F_output_5F_0.bin(float32:)*output:_2F_audioEncoder_2F_Mul_5F_1_5F_output_5F_0(0,512,0)
/audioEncoder/conv2/Conv:Conv*input:_2F_audioEncoder_2F_Mul_5F_1_5F_output_5F_0(0,512,0);onnx_3A__3A_Conv_5F_1220_nchw.bin(float32:512,512,3,1);onnx_3A__3A_Conv_5F_1221.bin(float32:512)*output:_2F_audioEncoder_2F_conv2_2F_Conv_5F_output_5F_0(0,512,0)*dilations:1;group:1;kernel_shape:3;pads:1,1;strides:2
/audioEncoder/Div_1:Div*input:_2F_audioEncoder_2F_conv2_2F_Conv_5F_output_5F_0(0,512,0);_2F_audioEncoder_2F_Constant_5F_output_5F_0.bin(float32:)*output:_2F_audioEncoder_2F_Div_5F_1_5F_output_5F_0(0,512,0)
/audioEncoder/Erf_1:Erf*input:_2F_audioEncoder_2F_Div_5F_1_5F_output_5F_0(0,512,0)*output:_2F_audioEncoder_2F_Erf_5F_1_5F_output_5F_0(0,512,0)
/audioEncoder/Add_1:Add*input:_2F_audioEncoder_2F_Erf_5F_1_5F_output_5F_0(0,512,0);_2F_audioEncoder_2F_Constant_5F_1_5F_output_5F_0.bin(float32:)*output:_2F_audioEncoder_2F_Add_5F_1_5F_output_5F_0(0,512,0)
/audioEncoder/Mul_2:Mul*input:_2F_audioEncoder_2F_conv2_2F_Conv_5F_output_5F_0(0,512,0);_2F_audioEncoder_2F_Add_5F_1_5F_output_5F_0(0,512,0)*output:_2F_audioEncoder_2F_Mul_5F_2_5F_output_5F_0(0,512,0)
/audioEncoder/Mul_3:Mul*input:_2F_audioEncoder_2F_Mul_5F_2_5F_output_5F_0(0,512,0);_2F_audioEncoder_2F_Constant_5F_2_5F_output_5F_0.bin(float32:)*output:_2F_audioEncoder_2F_Mul_5F_3_5F_output_5F_0(0,512,0)
/audioEncoder/Transpose:Transpose*input:_2F_audioEncoder_2F_Mul_5F_3_5F_output_5F_0(0,512,0)*output:_2F_audioEncoder_2F_Transpose_5F_output_5F_0(0,0,512)*perm:0,2,1
/audioEncoder/Shape:Shape*input:_2F_audioEncoder_2F_Transpose_5F_output_5F_0(0,0,512)*output:_2F_audioEncoder_2F_Shape_5F_output_5F_0(3)
/audioEncoder/Gather:Gather*input:_2F_audioEncoder_2F_Shape_5F_output_5F_0(3);_2F_audioEncoder_2F_Constant_5F_6_5F_output_5F_0.bin(int64:)*output:_2F_audioEncoder_2F_Gather_5F_output_5F_0()*axis:0
/audioEncoder/Unsqueeze:Unsqueeze*input:_2F_audioEncoder_2F_Gather_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_Unsqueeze_5F_output_5F_0(1)
/audioEncoder/Slice:Slice*input:audioEncoder_2E_positional_5F_embedding.bin(float32:1500,512);_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1);_2F_audioEncoder_2F_Unsqueeze_5F_output_5F_0(1);_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1);_2F_audioEncoder_2F_Constant_5F_10_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_Slice_5F_output_5F_0(0,0)
/audioEncoder/Add_2:Add*input:_2F_audioEncoder_2F_Transpose_5F_output_5F_0(0,0,512);_2F_audioEncoder_2F_Slice_5F_output_5F_0(0,0)*output:_2F_audioEncoder_2F_Add_5F_2_5F_output_5F_0(0,0,512)
/audioEncoder/blocks.0/attn_ln/ReduceMean:ReduceMean*input:_2F_audioEncoder_2F_Add_5F_2_5F_output_5F_0(0,0,512)*output:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_5F_ln_2F_ReduceMean_5F_output_5F_0(0,0,1)*axes:-1
/audioEncoder/blocks.0/attn_ln/Sub:Sub*input:_2F_audioEncoder_2F_Add_5F_2_5F_output_5F_0(0,0,512);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_5F_ln_2F_ReduceMean_5F_output_5F_0(0,0,1)*output:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_5F_ln_2F_Sub_5F_output_5F_0(0,0,512)
/audioEncoder/blocks.0/attn_ln/Pow:Pow*input:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_5F_ln_2F_Sub_5F_output_5F_0(0,0,512);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_5F_ln_2F_Constant_5F_output_5F_0.bin(float32:)*output:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_5F_ln_2F_Pow_5F_output_5F_0(0,0,512)
/audioEncoder/blocks.0/attn_ln/ReduceMean_1:ReduceMean*input:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_5F_ln_2F_Pow_5F_output_5F_0(0,0,512)*output:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_5F_ln_2F_ReduceMean_5F_1_5F_output_5F_0(0,0,1)*axes:-1
/audioEncoder/blocks.0/attn_ln/Add:Add*input:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_5F_ln_2F_ReduceMean_5F_1_5F_output_5F_0(0,0,1);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_5F_ln_2F_Constant_5F_1_5F_output_5F_0.bin(float32:)*output:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_5F_ln_2F_Add_5F_output_5F_0(0,0,1)
/audioEncoder/blocks.0/attn_ln/Sqrt:Sqrt*input:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_5F_ln_2F_Add_5F_output_5F_0(0,0,1)*output:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_5F_ln_2F_Sqrt_5F_output_5F_0(0,0,1)
/audioEncoder/blocks.0/attn_ln/Div:Div*input:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_5F_ln_2F_Sub_5F_output_5F_0(0,0,512);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_5F_ln_2F_Sqrt_5F_output_5F_0(0,0,1)*output:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_5F_ln_2F_Div_5F_output_5F_0(0,0,512)
/audioEncoder/blocks.0/attn_ln/Mul:Mul*input:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_5F_ln_2F_Div_5F_output_5F_0(0,0,512);audioEncoder_2E_blocks_2E_0_2E_attn_5F_ln_2E_weight.bin(float32:512)*output:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_5F_ln_2F_Mul_5F_output_5F_0(0,0,512)
/audioEncoder/blocks.0/attn_ln/Add_1:Add*input:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_5F_ln_2F_Mul_5F_output_5F_0(0,0,512);audioEncoder_2E_blocks_2E_0_2E_attn_5F_ln_2E_bias.bin(float32:512)*output:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_5F_ln_2F_Add_5F_1_5F_output_5F_0(0,0,512)
/audioEncoder/blocks.0/attn/query/MatMul:MatMul*input:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_5F_ln_2F_Add_5F_1_5F_output_5F_0(0,0,512);onnx_3A__3A_MatMul_5F_1226.bin(float32:512,512)*output:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_query_2F_MatMul_5F_output_5F_0(0,0,512)
/audioEncoder/blocks.0/attn/query/Add:Add*input:onnx_3A__3A_Add_5F_1225.bin(float32:512);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_query_2F_MatMul_5F_output_5F_0(0,0,512)*output:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_query_2F_Add_5F_output_5F_0(0,0,512)
/audioEncoder/blocks.0/attn/key/MatMul:MatMul*input:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_5F_ln_2F_Add_5F_1_5F_output_5F_0(0,0,512);onnx_3A__3A_MatMul_5F_1228.bin(float32:512,512)*output:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_key_2F_MatMul_5F_output_5F_0(0,0,512)
/audioEncoder/blocks.0/attn/value/MatMul:MatMul*input:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_5F_ln_2F_Add_5F_1_5F_output_5F_0(0,0,512);onnx_3A__3A_MatMul_5F_1231.bin(float32:512,512)*output:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_value_2F_MatMul_5F_output_5F_0(0,0,512)
/audioEncoder/blocks.0/attn/value/Add:Add*input:onnx_3A__3A_Add_5F_1230.bin(float32:512);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_value_2F_MatMul_5F_output_5F_0(0,0,512)*output:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_value_2F_Add_5F_output_5F_0(0,0,512)
/audioEncoder/blocks.0/attn/Shape_1:Shape*input:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_query_2F_Add_5F_output_5F_0(0,0,512)*output:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Shape_5F_1_5F_output_5F_0(3)
/audioEncoder/blocks.0/attn/Gather_1:Gather*input:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Shape_5F_1_5F_output_5F_0(3);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_3_5F_output_5F_0.bin(int64:)*output:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Gather_5F_1_5F_output_5F_0()*axis:0
/audioEncoder/blocks.0/attn/Gather_2:Gather*input:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Shape_5F_1_5F_output_5F_0(3);_2F_audioEncoder_2F_Constant_5F_6_5F_output_5F_0.bin(int64:)*output:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Gather_5F_2_5F_output_5F_0()*axis:0
/audioEncoder/blocks.0/attn/Unsqueeze:Unsqueeze*input:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Gather_5F_1_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Unsqueeze_5F_output_5F_0(1)
/audioEncoder/blocks.0/attn/Unsqueeze_1:Unsqueeze*input:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Gather_5F_2_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Unsqueeze_5F_1_5F_output_5F_0(1)
/audioEncoder/blocks.0/attn/Concat:Concat*input:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Unsqueeze_5F_output_5F_0(1);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Unsqueeze_5F_1_5F_output_5F_0(1);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_5_5F_output_5F_0.bin(int64:1);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_6_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Concat_5F_output_5F_0(4)*axis:0
/audioEncoder/blocks.0/attn/Reshape:Reshape*input:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_query_2F_Add_5F_output_5F_0(0,0,512);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Concat_5F_output_5F_0(4)*output:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Reshape_5F_output_5F_0()
/audioEncoder/blocks.0/attn/Transpose:Transpose*input:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Reshape_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Transpose_5F_output_5F_0()*perm:0,2,1,3
/audioEncoder/blocks.0/attn/Mul:Mul*input:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Transpose_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Pow_5F_output_5F_0.bin(float32:)*output:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Mul_5F_output_5F_0()
/audioEncoder/blocks.0/attn/Shape_3:Shape*input:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_key_2F_MatMul_5F_output_5F_0(0,0,512)*output:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Shape_5F_3_5F_output_5F_0(3)
/audioEncoder/blocks.0/attn/Gather_3:Gather*input:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Shape_5F_3_5F_output_5F_0(3);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_3_5F_output_5F_0.bin(int64:)*output:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Gather_5F_3_5F_output_5F_0()*axis:0
/audioEncoder/blocks.0/attn/Gather_4:Gather*input:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Shape_5F_3_5F_output_5F_0(3);_2F_audioEncoder_2F_Constant_5F_6_5F_output_5F_0.bin(int64:)*output:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Gather_5F_4_5F_output_5F_0()*axis:0
/audioEncoder/blocks.0/attn/Unsqueeze_2:Unsqueeze*input:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Gather_5F_3_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Unsqueeze_5F_2_5F_output_5F_0(1)
/audioEncoder/blocks.0/attn/Unsqueeze_3:Unsqueeze*input:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Gather_5F_4_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Unsqueeze_5F_3_5F_output_5F_0(1)
/audioEncoder/blocks.0/attn/Concat_1:Concat*input:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Unsqueeze_5F_2_5F_output_5F_0(1);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Unsqueeze_5F_3_5F_output_5F_0(1);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_5_5F_output_5F_0.bin(int64:1);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_6_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Concat_5F_1_5F_output_5F_0(4)*axis:0
/audioEncoder/blocks.0/attn/Reshape_1:Reshape*input:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_key_2F_MatMul_5F_output_5F_0(0,0,512);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Concat_5F_1_5F_output_5F_0(4)*output:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Reshape_5F_1_5F_output_5F_0()
/audioEncoder/blocks.0/attn/Transpose_1:Transpose*input:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Reshape_5F_1_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Transpose_5F_1_5F_output_5F_0()*perm:0,2,3,1
/audioEncoder/blocks.0/attn/Mul_1:Mul*input:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Transpose_5F_1_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Pow_5F_output_5F_0.bin(float32:)*output:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Mul_5F_1_5F_output_5F_0()
/audioEncoder/blocks.0/attn/Shape_5:Shape*input:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_value_2F_Add_5F_output_5F_0(0,0,512)*output:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Shape_5F_5_5F_output_5F_0(3)
/audioEncoder/blocks.0/attn/Gather_5:Gather*input:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Shape_5F_5_5F_output_5F_0(3);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_3_5F_output_5F_0.bin(int64:)*output:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Gather_5F_5_5F_output_5F_0()*axis:0
/audioEncoder/blocks.0/attn/Gather_6:Gather*input:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Shape_5F_5_5F_output_5F_0(3);_2F_audioEncoder_2F_Constant_5F_6_5F_output_5F_0.bin(int64:)*output:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Gather_5F_6_5F_output_5F_0()*axis:0
/audioEncoder/blocks.0/attn/Unsqueeze_4:Unsqueeze*input:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Gather_5F_5_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Unsqueeze_5F_4_5F_output_5F_0(1)
/audioEncoder/blocks.0/attn/Unsqueeze_5:Unsqueeze*input:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Gather_5F_6_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Unsqueeze_5F_5_5F_output_5F_0(1)
/audioEncoder/blocks.0/attn/Concat_2:Concat*input:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Unsqueeze_5F_4_5F_output_5F_0(1);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Unsqueeze_5F_5_5F_output_5F_0(1);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_5_5F_output_5F_0.bin(int64:1);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_6_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Concat_5F_2_5F_output_5F_0(4)*axis:0
/audioEncoder/blocks.0/attn/Reshape_2:Reshape*input:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_value_2F_Add_5F_output_5F_0(0,0,512);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Concat_5F_2_5F_output_5F_0(4)*output:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Reshape_5F_2_5F_output_5F_0()
/audioEncoder/blocks.0/attn/Transpose_2:Transpose*input:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Reshape_5F_2_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Transpose_5F_2_5F_output_5F_0()*perm:0,2,1,3
/audioEncoder/blocks.0/attn/MatMul:MatMul*input:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Mul_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Mul_5F_1_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_MatMul_5F_output_5F_0()
/audioEncoder/blocks.0/attn/Softmax:Softmax*input:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_MatMul_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Softmax_5F_output_5F_0()*axis:-1
/audioEncoder/blocks.0/attn/MatMul_1:MatMul*input:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Softmax_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Transpose_5F_2_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_MatMul_5F_1_5F_output_5F_0()
/audioEncoder/blocks.0/attn/Transpose_3:Transpose*input:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_MatMul_5F_1_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Transpose_5F_3_5F_output_5F_0()*perm:0,2,1,3
/audioEncoder/blocks.0/attn/Shape_7:Shape*input:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Transpose_5F_3_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Shape_5F_7_5F_output_5F_0(0)
/audioEncoder/blocks.0/attn/Slice:Slice*input:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Shape_5F_7_5F_output_5F_0(0);_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_17_5F_output_5F_0.bin(int64:1);_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Slice_5F_output_5F_0(0)
/audioEncoder/blocks.0/attn/Concat_3:Concat*input:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Slice_5F_output_5F_0(0);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_6_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Concat_5F_3_5F_output_5F_0(0)*axis:0
/audioEncoder/blocks.0/attn/Reshape_3:Reshape*input:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Transpose_5F_3_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Concat_5F_3_5F_output_5F_0(0)*output:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Reshape_5F_3_5F_output_5F_0()
/audioEncoder/blocks.0/attn/out/MatMul:MatMul*input:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Reshape_5F_3_5F_output_5F_0();onnx_3A__3A_MatMul_5F_1240.bin(float32:512,512)*output:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_out_2F_MatMul_5F_output_5F_0()
/audioEncoder/blocks.0/attn/out/Add:Add*input:onnx_3A__3A_Add_5F_1239.bin(float32:512);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_out_2F_MatMul_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_out_2F_Add_5F_output_5F_0()
/audioEncoder/blocks.0/Add:Add*input:_2F_audioEncoder_2F_Add_5F_2_5F_output_5F_0(0,0,512);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_out_2F_Add_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_0_2F_Add_5F_output_5F_0()
/audioEncoder/blocks.0/mlp_ln/ReduceMean:ReduceMean*input:_2F_audioEncoder_2F_blocks_2E_0_2F_Add_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_0_2F_mlp_5F_ln_2F_ReduceMean_5F_output_5F_0()*axes:-1
/audioEncoder/blocks.0/mlp_ln/Sub:Sub*input:_2F_audioEncoder_2F_blocks_2E_0_2F_Add_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_0_2F_mlp_5F_ln_2F_ReduceMean_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_0_2F_mlp_5F_ln_2F_Sub_5F_output_5F_0()
/audioEncoder/blocks.0/mlp_ln/Pow:Pow*input:_2F_audioEncoder_2F_blocks_2E_0_2F_mlp_5F_ln_2F_Sub_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_0_2F_attn_5F_ln_2F_Constant_5F_output_5F_0.bin(float32:)*output:_2F_audioEncoder_2F_blocks_2E_0_2F_mlp_5F_ln_2F_Pow_5F_output_5F_0()
/audioEncoder/blocks.0/mlp_ln/ReduceMean_1:ReduceMean*input:_2F_audioEncoder_2F_blocks_2E_0_2F_mlp_5F_ln_2F_Pow_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_0_2F_mlp_5F_ln_2F_ReduceMean_5F_1_5F_output_5F_0()*axes:-1
/audioEncoder/blocks.0/mlp_ln/Add:Add*input:_2F_audioEncoder_2F_blocks_2E_0_2F_mlp_5F_ln_2F_ReduceMean_5F_1_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_0_2F_attn_5F_ln_2F_Constant_5F_1_5F_output_5F_0.bin(float32:)*output:_2F_audioEncoder_2F_blocks_2E_0_2F_mlp_5F_ln_2F_Add_5F_output_5F_0()
/audioEncoder/blocks.0/mlp_ln/Sqrt:Sqrt*input:_2F_audioEncoder_2F_blocks_2E_0_2F_mlp_5F_ln_2F_Add_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_0_2F_mlp_5F_ln_2F_Sqrt_5F_output_5F_0()
/audioEncoder/blocks.0/mlp_ln/Div:Div*input:_2F_audioEncoder_2F_blocks_2E_0_2F_mlp_5F_ln_2F_Sub_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_0_2F_mlp_5F_ln_2F_Sqrt_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_0_2F_mlp_5F_ln_2F_Div_5F_output_5F_0()
/audioEncoder/blocks.0/mlp_ln/Mul:Mul*input:_2F_audioEncoder_2F_blocks_2E_0_2F_mlp_5F_ln_2F_Div_5F_output_5F_0();audioEncoder_2E_blocks_2E_0_2E_mlp_5F_ln_2E_weight.bin(float32:512)*output:_2F_audioEncoder_2F_blocks_2E_0_2F_mlp_5F_ln_2F_Mul_5F_output_5F_0()
/audioEncoder/blocks.0/mlp_ln/Add_1:Add*input:_2F_audioEncoder_2F_blocks_2E_0_2F_mlp_5F_ln_2F_Mul_5F_output_5F_0();audioEncoder_2E_blocks_2E_0_2E_mlp_5F_ln_2E_bias.bin(float32:512)*output:_2F_audioEncoder_2F_blocks_2E_0_2F_mlp_5F_ln_2F_Add_5F_1_5F_output_5F_0()
/audioEncoder/blocks.0/mlp/mlp.0/MatMul:MatMul*input:_2F_audioEncoder_2F_blocks_2E_0_2F_mlp_5F_ln_2F_Add_5F_1_5F_output_5F_0();onnx_3A__3A_MatMul_5F_1243.bin(float32:512,2048)*output:_2F_audioEncoder_2F_blocks_2E_0_2F_mlp_2F_mlp_2E_0_2F_MatMul_5F_output_5F_0()
/audioEncoder/blocks.0/mlp/mlp.0/Add:Add*input:onnx_3A__3A_Add_5F_1242.bin(float32:2048);_2F_audioEncoder_2F_blocks_2E_0_2F_mlp_2F_mlp_2E_0_2F_MatMul_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_0_2F_mlp_2F_mlp_2E_0_2F_Add_5F_output_5F_0()
/audioEncoder/blocks.0/mlp/mlp.1/Div:Div*input:_2F_audioEncoder_2F_blocks_2E_0_2F_mlp_2F_mlp_2E_0_2F_Add_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_output_5F_0.bin(float32:)*output:_2F_audioEncoder_2F_blocks_2E_0_2F_mlp_2F_mlp_2E_1_2F_Div_5F_output_5F_0()
/audioEncoder/blocks.0/mlp/mlp.1/Erf:Erf*input:_2F_audioEncoder_2F_blocks_2E_0_2F_mlp_2F_mlp_2E_1_2F_Div_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_0_2F_mlp_2F_mlp_2E_1_2F_Erf_5F_output_5F_0()
/audioEncoder/blocks.0/mlp/mlp.1/Add:Add*input:_2F_audioEncoder_2F_blocks_2E_0_2F_mlp_2F_mlp_2E_1_2F_Erf_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_1_5F_output_5F_0.bin(float32:)*output:_2F_audioEncoder_2F_blocks_2E_0_2F_mlp_2F_mlp_2E_1_2F_Add_5F_output_5F_0()
/audioEncoder/blocks.0/mlp/mlp.1/Mul:Mul*input:_2F_audioEncoder_2F_blocks_2E_0_2F_mlp_2F_mlp_2E_0_2F_Add_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_0_2F_mlp_2F_mlp_2E_1_2F_Add_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_0_2F_mlp_2F_mlp_2E_1_2F_Mul_5F_output_5F_0()
/audioEncoder/blocks.0/mlp/mlp.1/Mul_1:Mul*input:_2F_audioEncoder_2F_blocks_2E_0_2F_mlp_2F_mlp_2E_1_2F_Mul_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_2_5F_output_5F_0.bin(float32:)*output:_2F_audioEncoder_2F_blocks_2E_0_2F_mlp_2F_mlp_2E_1_2F_Mul_5F_1_5F_output_5F_0()
/audioEncoder/blocks.0/mlp/mlp.2/MatMul:MatMul*input:_2F_audioEncoder_2F_blocks_2E_0_2F_mlp_2F_mlp_2E_1_2F_Mul_5F_1_5F_output_5F_0();onnx_3A__3A_MatMul_5F_1246.bin(float32:2048,512)*output:_2F_audioEncoder_2F_blocks_2E_0_2F_mlp_2F_mlp_2E_2_2F_MatMul_5F_output_5F_0()
/audioEncoder/blocks.0/mlp/mlp.2/Add:Add*input:onnx_3A__3A_Add_5F_1245.bin(float32:512);_2F_audioEncoder_2F_blocks_2E_0_2F_mlp_2F_mlp_2E_2_2F_MatMul_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_0_2F_mlp_2F_mlp_2E_2_2F_Add_5F_output_5F_0()
/audioEncoder/blocks.0/Add_1:Add*input:_2F_audioEncoder_2F_blocks_2E_0_2F_Add_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_0_2F_mlp_2F_mlp_2E_2_2F_Add_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_0_2F_Add_5F_1_5F_output_5F_0()
/audioEncoder/blocks.1/attn_ln/ReduceMean:ReduceMean*input:_2F_audioEncoder_2F_blocks_2E_0_2F_Add_5F_1_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_5F_ln_2F_ReduceMean_5F_output_5F_0()*axes:-1
/audioEncoder/blocks.1/attn_ln/Sub:Sub*input:_2F_audioEncoder_2F_blocks_2E_0_2F_Add_5F_1_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_1_2F_attn_5F_ln_2F_ReduceMean_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_5F_ln_2F_Sub_5F_output_5F_0()
/audioEncoder/blocks.1/attn_ln/Pow:Pow*input:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_5F_ln_2F_Sub_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_0_2F_attn_5F_ln_2F_Constant_5F_output_5F_0.bin(float32:)*output:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_5F_ln_2F_Pow_5F_output_5F_0()
/audioEncoder/blocks.1/attn_ln/ReduceMean_1:ReduceMean*input:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_5F_ln_2F_Pow_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_5F_ln_2F_ReduceMean_5F_1_5F_output_5F_0()*axes:-1
/audioEncoder/blocks.1/attn_ln/Add:Add*input:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_5F_ln_2F_ReduceMean_5F_1_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_0_2F_attn_5F_ln_2F_Constant_5F_1_5F_output_5F_0.bin(float32:)*output:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_5F_ln_2F_Add_5F_output_5F_0()
/audioEncoder/blocks.1/attn_ln/Sqrt:Sqrt*input:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_5F_ln_2F_Add_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_5F_ln_2F_Sqrt_5F_output_5F_0()
/audioEncoder/blocks.1/attn_ln/Div:Div*input:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_5F_ln_2F_Sub_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_1_2F_attn_5F_ln_2F_Sqrt_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_5F_ln_2F_Div_5F_output_5F_0()
/audioEncoder/blocks.1/attn_ln/Mul:Mul*input:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_5F_ln_2F_Div_5F_output_5F_0();audioEncoder_2E_blocks_2E_1_2E_attn_5F_ln_2E_weight.bin(float32:512)*output:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_5F_ln_2F_Mul_5F_output_5F_0()
/audioEncoder/blocks.1/attn_ln/Add_1:Add*input:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_5F_ln_2F_Mul_5F_output_5F_0();audioEncoder_2E_blocks_2E_1_2E_attn_5F_ln_2E_bias.bin(float32:512)*output:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_5F_ln_2F_Add_5F_1_5F_output_5F_0()
/audioEncoder/blocks.1/attn/query/MatMul:MatMul*input:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_5F_ln_2F_Add_5F_1_5F_output_5F_0();onnx_3A__3A_MatMul_5F_1249.bin(float32:512,512)*output:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_query_2F_MatMul_5F_output_5F_0()
/audioEncoder/blocks.1/attn/query/Add:Add*input:onnx_3A__3A_Add_5F_1248.bin(float32:512);_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_query_2F_MatMul_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_query_2F_Add_5F_output_5F_0()
/audioEncoder/blocks.1/attn/key/MatMul:MatMul*input:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_5F_ln_2F_Add_5F_1_5F_output_5F_0();onnx_3A__3A_MatMul_5F_1251.bin(float32:512,512)*output:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_key_2F_MatMul_5F_output_5F_0()
/audioEncoder/blocks.1/attn/value/MatMul:MatMul*input:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_5F_ln_2F_Add_5F_1_5F_output_5F_0();onnx_3A__3A_MatMul_5F_1254.bin(float32:512,512)*output:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_value_2F_MatMul_5F_output_5F_0()
/audioEncoder/blocks.1/attn/value/Add:Add*input:onnx_3A__3A_Add_5F_1253.bin(float32:512);_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_value_2F_MatMul_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_value_2F_Add_5F_output_5F_0()
/audioEncoder/blocks.1/attn/Shape:Shape*input:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_query_2F_Add_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Shape_5F_output_5F_0(0)
/audioEncoder/blocks.1/attn/Gather:Gather*input:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Shape_5F_output_5F_0(0);_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Constant_5F_output_5F_0.bin(int64:)*output:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Gather_5F_output_5F_0()*axis:0
/audioEncoder/blocks.1/attn/Div:Div*input:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Gather_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_1_5F_output_5F_0.bin(int64:)*output:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Div_5F_output_5F_0()
/audioEncoder/blocks.1/attn/Cast_2:Cast*input:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Div_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Cast_5F_2_5F_output_5F_0()*to:1
/audioEncoder/blocks.1/attn/Pow:Pow*input:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Cast_5F_2_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_2_5F_output_5F_0.bin(float32:)*output:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Pow_5F_output_5F_0()
/audioEncoder/blocks.1/attn/Gather_1:Gather*input:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Shape_5F_output_5F_0(0);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_3_5F_output_5F_0.bin(int64:)*output:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Gather_5F_1_5F_output_5F_0()*axis:0
/audioEncoder/blocks.1/attn/Gather_2:Gather*input:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Shape_5F_output_5F_0(0);_2F_audioEncoder_2F_Constant_5F_6_5F_output_5F_0.bin(int64:)*output:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Gather_5F_2_5F_output_5F_0()*axis:0
/audioEncoder/blocks.1/attn/Unsqueeze:Unsqueeze*input:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Gather_5F_1_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Unsqueeze_5F_output_5F_0(1)
/audioEncoder/blocks.1/attn/Unsqueeze_1:Unsqueeze*input:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Gather_5F_2_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Unsqueeze_5F_1_5F_output_5F_0(1)
/audioEncoder/blocks.1/attn/Concat:Concat*input:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Unsqueeze_5F_output_5F_0(1);_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Unsqueeze_5F_1_5F_output_5F_0(1);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_5_5F_output_5F_0.bin(int64:1);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_6_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Concat_5F_output_5F_0(4)*axis:0
/audioEncoder/blocks.1/attn/Reshape:Reshape*input:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_query_2F_Add_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Concat_5F_output_5F_0(4)*output:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Reshape_5F_output_5F_0()
/audioEncoder/blocks.1/attn/Transpose:Transpose*input:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Reshape_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Transpose_5F_output_5F_0()*perm:0,2,1,3
/audioEncoder/blocks.1/attn/Mul:Mul*input:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Transpose_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Pow_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Mul_5F_output_5F_0()
/audioEncoder/blocks.1/attn/Shape_3:Shape*input:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_key_2F_MatMul_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Shape_5F_3_5F_output_5F_0(0)
/audioEncoder/blocks.1/attn/Gather_3:Gather*input:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Shape_5F_3_5F_output_5F_0(0);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_3_5F_output_5F_0.bin(int64:)*output:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Gather_5F_3_5F_output_5F_0()*axis:0
/audioEncoder/blocks.1/attn/Gather_4:Gather*input:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Shape_5F_3_5F_output_5F_0(0);_2F_audioEncoder_2F_Constant_5F_6_5F_output_5F_0.bin(int64:)*output:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Gather_5F_4_5F_output_5F_0()*axis:0
/audioEncoder/blocks.1/attn/Unsqueeze_2:Unsqueeze*input:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Gather_5F_3_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Unsqueeze_5F_2_5F_output_5F_0(1)
/audioEncoder/blocks.1/attn/Unsqueeze_3:Unsqueeze*input:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Gather_5F_4_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Unsqueeze_5F_3_5F_output_5F_0(1)
/audioEncoder/blocks.1/attn/Concat_1:Concat*input:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Unsqueeze_5F_2_5F_output_5F_0(1);_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Unsqueeze_5F_3_5F_output_5F_0(1);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_5_5F_output_5F_0.bin(int64:1);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_6_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Concat_5F_1_5F_output_5F_0(4)*axis:0
/audioEncoder/blocks.1/attn/Reshape_1:Reshape*input:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_key_2F_MatMul_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Concat_5F_1_5F_output_5F_0(4)*output:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Reshape_5F_1_5F_output_5F_0()
/audioEncoder/blocks.1/attn/Transpose_1:Transpose*input:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Reshape_5F_1_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Transpose_5F_1_5F_output_5F_0()*perm:0,2,3,1
/audioEncoder/blocks.1/attn/Mul_1:Mul*input:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Transpose_5F_1_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Pow_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Mul_5F_1_5F_output_5F_0()
/audioEncoder/blocks.1/attn/Shape_5:Shape*input:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_value_2F_Add_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Shape_5F_5_5F_output_5F_0(0)
/audioEncoder/blocks.1/attn/Gather_5:Gather*input:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Shape_5F_5_5F_output_5F_0(0);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_3_5F_output_5F_0.bin(int64:)*output:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Gather_5F_5_5F_output_5F_0()*axis:0
/audioEncoder/blocks.1/attn/Gather_6:Gather*input:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Shape_5F_5_5F_output_5F_0(0);_2F_audioEncoder_2F_Constant_5F_6_5F_output_5F_0.bin(int64:)*output:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Gather_5F_6_5F_output_5F_0()*axis:0
/audioEncoder/blocks.1/attn/Unsqueeze_4:Unsqueeze*input:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Gather_5F_5_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Unsqueeze_5F_4_5F_output_5F_0(1)
/audioEncoder/blocks.1/attn/Unsqueeze_5:Unsqueeze*input:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Gather_5F_6_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Unsqueeze_5F_5_5F_output_5F_0(1)
/audioEncoder/blocks.1/attn/Concat_2:Concat*input:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Unsqueeze_5F_4_5F_output_5F_0(1);_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Unsqueeze_5F_5_5F_output_5F_0(1);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_5_5F_output_5F_0.bin(int64:1);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_6_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Concat_5F_2_5F_output_5F_0(4)*axis:0
/audioEncoder/blocks.1/attn/Reshape_2:Reshape*input:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_value_2F_Add_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Concat_5F_2_5F_output_5F_0(4)*output:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Reshape_5F_2_5F_output_5F_0()
/audioEncoder/blocks.1/attn/Transpose_2:Transpose*input:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Reshape_5F_2_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Transpose_5F_2_5F_output_5F_0()*perm:0,2,1,3
/audioEncoder/blocks.1/attn/MatMul:MatMul*input:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Mul_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Mul_5F_1_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_MatMul_5F_output_5F_0()
/audioEncoder/blocks.1/attn/Softmax:Softmax*input:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_MatMul_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Softmax_5F_output_5F_0()*axis:-1
/audioEncoder/blocks.1/attn/MatMul_1:MatMul*input:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Softmax_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Transpose_5F_2_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_MatMul_5F_1_5F_output_5F_0()
/audioEncoder/blocks.1/attn/Transpose_3:Transpose*input:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_MatMul_5F_1_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Transpose_5F_3_5F_output_5F_0()*perm:0,2,1,3
/audioEncoder/blocks.1/attn/Shape_7:Shape*input:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Transpose_5F_3_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Shape_5F_7_5F_output_5F_0(0)
/audioEncoder/blocks.1/attn/Slice:Slice*input:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Shape_5F_7_5F_output_5F_0(0);_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_17_5F_output_5F_0.bin(int64:1);_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Slice_5F_output_5F_0(0)
/audioEncoder/blocks.1/attn/Concat_3:Concat*input:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Slice_5F_output_5F_0(0);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_6_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Concat_5F_3_5F_output_5F_0(0)*axis:0
/audioEncoder/blocks.1/attn/Reshape_3:Reshape*input:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Transpose_5F_3_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Concat_5F_3_5F_output_5F_0(0)*output:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Reshape_5F_3_5F_output_5F_0()
/audioEncoder/blocks.1/attn/out/MatMul:MatMul*input:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Reshape_5F_3_5F_output_5F_0();onnx_3A__3A_MatMul_5F_1263.bin(float32:512,512)*output:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_out_2F_MatMul_5F_output_5F_0()
/audioEncoder/blocks.1/attn/out/Add:Add*input:onnx_3A__3A_Add_5F_1262.bin(float32:512);_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_out_2F_MatMul_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_out_2F_Add_5F_output_5F_0()
/audioEncoder/blocks.1/Add:Add*input:_2F_audioEncoder_2F_blocks_2E_0_2F_Add_5F_1_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_out_2F_Add_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_1_2F_Add_5F_output_5F_0()
/audioEncoder/blocks.1/mlp_ln/ReduceMean:ReduceMean*input:_2F_audioEncoder_2F_blocks_2E_1_2F_Add_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_1_2F_mlp_5F_ln_2F_ReduceMean_5F_output_5F_0()*axes:-1
/audioEncoder/blocks.1/mlp_ln/Sub:Sub*input:_2F_audioEncoder_2F_blocks_2E_1_2F_Add_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_1_2F_mlp_5F_ln_2F_ReduceMean_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_1_2F_mlp_5F_ln_2F_Sub_5F_output_5F_0()
/audioEncoder/blocks.1/mlp_ln/Pow:Pow*input:_2F_audioEncoder_2F_blocks_2E_1_2F_mlp_5F_ln_2F_Sub_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_0_2F_attn_5F_ln_2F_Constant_5F_output_5F_0.bin(float32:)*output:_2F_audioEncoder_2F_blocks_2E_1_2F_mlp_5F_ln_2F_Pow_5F_output_5F_0()
/audioEncoder/blocks.1/mlp_ln/ReduceMean_1:ReduceMean*input:_2F_audioEncoder_2F_blocks_2E_1_2F_mlp_5F_ln_2F_Pow_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_1_2F_mlp_5F_ln_2F_ReduceMean_5F_1_5F_output_5F_0()*axes:-1
/audioEncoder/blocks.1/mlp_ln/Add:Add*input:_2F_audioEncoder_2F_blocks_2E_1_2F_mlp_5F_ln_2F_ReduceMean_5F_1_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_0_2F_attn_5F_ln_2F_Constant_5F_1_5F_output_5F_0.bin(float32:)*output:_2F_audioEncoder_2F_blocks_2E_1_2F_mlp_5F_ln_2F_Add_5F_output_5F_0()
/audioEncoder/blocks.1/mlp_ln/Sqrt:Sqrt*input:_2F_audioEncoder_2F_blocks_2E_1_2F_mlp_5F_ln_2F_Add_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_1_2F_mlp_5F_ln_2F_Sqrt_5F_output_5F_0()
/audioEncoder/blocks.1/mlp_ln/Div:Div*input:_2F_audioEncoder_2F_blocks_2E_1_2F_mlp_5F_ln_2F_Sub_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_1_2F_mlp_5F_ln_2F_Sqrt_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_1_2F_mlp_5F_ln_2F_Div_5F_output_5F_0()
/audioEncoder/blocks.1/mlp_ln/Mul:Mul*input:_2F_audioEncoder_2F_blocks_2E_1_2F_mlp_5F_ln_2F_Div_5F_output_5F_0();audioEncoder_2E_blocks_2E_1_2E_mlp_5F_ln_2E_weight.bin(float32:512)*output:_2F_audioEncoder_2F_blocks_2E_1_2F_mlp_5F_ln_2F_Mul_5F_output_5F_0()
/audioEncoder/blocks.1/mlp_ln/Add_1:Add*input:_2F_audioEncoder_2F_blocks_2E_1_2F_mlp_5F_ln_2F_Mul_5F_output_5F_0();audioEncoder_2E_blocks_2E_1_2E_mlp_5F_ln_2E_bias.bin(float32:512)*output:_2F_audioEncoder_2F_blocks_2E_1_2F_mlp_5F_ln_2F_Add_5F_1_5F_output_5F_0()
/audioEncoder/blocks.1/mlp/mlp.0/MatMul:MatMul*input:_2F_audioEncoder_2F_blocks_2E_1_2F_mlp_5F_ln_2F_Add_5F_1_5F_output_5F_0();onnx_3A__3A_MatMul_5F_1266.bin(float32:512,2048)*output:_2F_audioEncoder_2F_blocks_2E_1_2F_mlp_2F_mlp_2E_0_2F_MatMul_5F_output_5F_0()
/audioEncoder/blocks.1/mlp/mlp.0/Add:Add*input:onnx_3A__3A_Add_5F_1265.bin(float32:2048);_2F_audioEncoder_2F_blocks_2E_1_2F_mlp_2F_mlp_2E_0_2F_MatMul_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_1_2F_mlp_2F_mlp_2E_0_2F_Add_5F_output_5F_0()
/audioEncoder/blocks.1/mlp/mlp.1/Div:Div*input:_2F_audioEncoder_2F_blocks_2E_1_2F_mlp_2F_mlp_2E_0_2F_Add_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_output_5F_0.bin(float32:)*output:_2F_audioEncoder_2F_blocks_2E_1_2F_mlp_2F_mlp_2E_1_2F_Div_5F_output_5F_0()
/audioEncoder/blocks.1/mlp/mlp.1/Erf:Erf*input:_2F_audioEncoder_2F_blocks_2E_1_2F_mlp_2F_mlp_2E_1_2F_Div_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_1_2F_mlp_2F_mlp_2E_1_2F_Erf_5F_output_5F_0()
/audioEncoder/blocks.1/mlp/mlp.1/Add:Add*input:_2F_audioEncoder_2F_blocks_2E_1_2F_mlp_2F_mlp_2E_1_2F_Erf_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_1_5F_output_5F_0.bin(float32:)*output:_2F_audioEncoder_2F_blocks_2E_1_2F_mlp_2F_mlp_2E_1_2F_Add_5F_output_5F_0()
/audioEncoder/blocks.1/mlp/mlp.1/Mul:Mul*input:_2F_audioEncoder_2F_blocks_2E_1_2F_mlp_2F_mlp_2E_0_2F_Add_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_1_2F_mlp_2F_mlp_2E_1_2F_Add_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_1_2F_mlp_2F_mlp_2E_1_2F_Mul_5F_output_5F_0()
/audioEncoder/blocks.1/mlp/mlp.1/Mul_1:Mul*input:_2F_audioEncoder_2F_blocks_2E_1_2F_mlp_2F_mlp_2E_1_2F_Mul_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_2_5F_output_5F_0.bin(float32:)*output:_2F_audioEncoder_2F_blocks_2E_1_2F_mlp_2F_mlp_2E_1_2F_Mul_5F_1_5F_output_5F_0()
/audioEncoder/blocks.1/mlp/mlp.2/MatMul:MatMul*input:_2F_audioEncoder_2F_blocks_2E_1_2F_mlp_2F_mlp_2E_1_2F_Mul_5F_1_5F_output_5F_0();onnx_3A__3A_MatMul_5F_1269.bin(float32:2048,512)*output:_2F_audioEncoder_2F_blocks_2E_1_2F_mlp_2F_mlp_2E_2_2F_MatMul_5F_output_5F_0()
/audioEncoder/blocks.1/mlp/mlp.2/Add:Add*input:onnx_3A__3A_Add_5F_1268.bin(float32:512);_2F_audioEncoder_2F_blocks_2E_1_2F_mlp_2F_mlp_2E_2_2F_MatMul_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_1_2F_mlp_2F_mlp_2E_2_2F_Add_5F_output_5F_0()
/audioEncoder/blocks.1/Add_1:Add*input:_2F_audioEncoder_2F_blocks_2E_1_2F_Add_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_1_2F_mlp_2F_mlp_2E_2_2F_Add_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_1_2F_Add_5F_1_5F_output_5F_0()
/audioEncoder/blocks.2/attn_ln/ReduceMean:ReduceMean*input:_2F_audioEncoder_2F_blocks_2E_1_2F_Add_5F_1_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_5F_ln_2F_ReduceMean_5F_output_5F_0()*axes:-1
/audioEncoder/blocks.2/attn_ln/Sub:Sub*input:_2F_audioEncoder_2F_blocks_2E_1_2F_Add_5F_1_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_2_2F_attn_5F_ln_2F_ReduceMean_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_5F_ln_2F_Sub_5F_output_5F_0()
/audioEncoder/blocks.2/attn_ln/Pow:Pow*input:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_5F_ln_2F_Sub_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_0_2F_attn_5F_ln_2F_Constant_5F_output_5F_0.bin(float32:)*output:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_5F_ln_2F_Pow_5F_output_5F_0()
/audioEncoder/blocks.2/attn_ln/ReduceMean_1:ReduceMean*input:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_5F_ln_2F_Pow_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_5F_ln_2F_ReduceMean_5F_1_5F_output_5F_0()*axes:-1
/audioEncoder/blocks.2/attn_ln/Add:Add*input:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_5F_ln_2F_ReduceMean_5F_1_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_0_2F_attn_5F_ln_2F_Constant_5F_1_5F_output_5F_0.bin(float32:)*output:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_5F_ln_2F_Add_5F_output_5F_0()
/audioEncoder/blocks.2/attn_ln/Sqrt:Sqrt*input:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_5F_ln_2F_Add_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_5F_ln_2F_Sqrt_5F_output_5F_0()
/audioEncoder/blocks.2/attn_ln/Div:Div*input:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_5F_ln_2F_Sub_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_2_2F_attn_5F_ln_2F_Sqrt_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_5F_ln_2F_Div_5F_output_5F_0()
/audioEncoder/blocks.2/attn_ln/Mul:Mul*input:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_5F_ln_2F_Div_5F_output_5F_0();audioEncoder_2E_blocks_2E_2_2E_attn_5F_ln_2E_weight.bin(float32:512)*output:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_5F_ln_2F_Mul_5F_output_5F_0()
/audioEncoder/blocks.2/attn_ln/Add_1:Add*input:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_5F_ln_2F_Mul_5F_output_5F_0();audioEncoder_2E_blocks_2E_2_2E_attn_5F_ln_2E_bias.bin(float32:512)*output:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_5F_ln_2F_Add_5F_1_5F_output_5F_0()
/audioEncoder/blocks.2/attn/query/MatMul:MatMul*input:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_5F_ln_2F_Add_5F_1_5F_output_5F_0();onnx_3A__3A_MatMul_5F_1272.bin(float32:512,512)*output:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_query_2F_MatMul_5F_output_5F_0()
/audioEncoder/blocks.2/attn/query/Add:Add*input:onnx_3A__3A_Add_5F_1271.bin(float32:512);_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_query_2F_MatMul_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_query_2F_Add_5F_output_5F_0()
/audioEncoder/blocks.2/attn/key/MatMul:MatMul*input:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_5F_ln_2F_Add_5F_1_5F_output_5F_0();onnx_3A__3A_MatMul_5F_1274.bin(float32:512,512)*output:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_key_2F_MatMul_5F_output_5F_0()
/audioEncoder/blocks.2/attn/value/MatMul:MatMul*input:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_5F_ln_2F_Add_5F_1_5F_output_5F_0();onnx_3A__3A_MatMul_5F_1277.bin(float32:512,512)*output:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_value_2F_MatMul_5F_output_5F_0()
/audioEncoder/blocks.2/attn/value/Add:Add*input:onnx_3A__3A_Add_5F_1276.bin(float32:512);_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_value_2F_MatMul_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_value_2F_Add_5F_output_5F_0()
/audioEncoder/blocks.2/attn/Shape:Shape*input:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_query_2F_Add_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Shape_5F_output_5F_0(0)
/audioEncoder/blocks.2/attn/Gather:Gather*input:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Shape_5F_output_5F_0(0);_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Constant_5F_output_5F_0.bin(int64:)*output:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Gather_5F_output_5F_0()*axis:0
/audioEncoder/blocks.2/attn/Div:Div*input:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Gather_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_1_5F_output_5F_0.bin(int64:)*output:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Div_5F_output_5F_0()
/audioEncoder/blocks.2/attn/Cast_2:Cast*input:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Div_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Cast_5F_2_5F_output_5F_0()*to:1
/audioEncoder/blocks.2/attn/Pow:Pow*input:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Cast_5F_2_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_2_5F_output_5F_0.bin(float32:)*output:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Pow_5F_output_5F_0()
/audioEncoder/blocks.2/attn/Gather_1:Gather*input:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Shape_5F_output_5F_0(0);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_3_5F_output_5F_0.bin(int64:)*output:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Gather_5F_1_5F_output_5F_0()*axis:0
/audioEncoder/blocks.2/attn/Gather_2:Gather*input:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Shape_5F_output_5F_0(0);_2F_audioEncoder_2F_Constant_5F_6_5F_output_5F_0.bin(int64:)*output:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Gather_5F_2_5F_output_5F_0()*axis:0
/audioEncoder/blocks.2/attn/Unsqueeze:Unsqueeze*input:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Gather_5F_1_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Unsqueeze_5F_output_5F_0(1)
/audioEncoder/blocks.2/attn/Unsqueeze_1:Unsqueeze*input:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Gather_5F_2_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Unsqueeze_5F_1_5F_output_5F_0(1)
/audioEncoder/blocks.2/attn/Concat:Concat*input:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Unsqueeze_5F_output_5F_0(1);_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Unsqueeze_5F_1_5F_output_5F_0(1);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_5_5F_output_5F_0.bin(int64:1);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_6_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Concat_5F_output_5F_0(4)*axis:0
/audioEncoder/blocks.2/attn/Reshape:Reshape*input:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_query_2F_Add_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Concat_5F_output_5F_0(4)*output:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Reshape_5F_output_5F_0()
/audioEncoder/blocks.2/attn/Transpose:Transpose*input:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Reshape_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Transpose_5F_output_5F_0()*perm:0,2,1,3
/audioEncoder/blocks.2/attn/Mul:Mul*input:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Transpose_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Pow_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Mul_5F_output_5F_0()
/audioEncoder/blocks.2/attn/Shape_3:Shape*input:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_key_2F_MatMul_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Shape_5F_3_5F_output_5F_0(0)
/audioEncoder/blocks.2/attn/Gather_3:Gather*input:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Shape_5F_3_5F_output_5F_0(0);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_3_5F_output_5F_0.bin(int64:)*output:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Gather_5F_3_5F_output_5F_0()*axis:0
/audioEncoder/blocks.2/attn/Gather_4:Gather*input:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Shape_5F_3_5F_output_5F_0(0);_2F_audioEncoder_2F_Constant_5F_6_5F_output_5F_0.bin(int64:)*output:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Gather_5F_4_5F_output_5F_0()*axis:0
/audioEncoder/blocks.2/attn/Unsqueeze_2:Unsqueeze*input:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Gather_5F_3_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Unsqueeze_5F_2_5F_output_5F_0(1)
/audioEncoder/blocks.2/attn/Unsqueeze_3:Unsqueeze*input:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Gather_5F_4_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Unsqueeze_5F_3_5F_output_5F_0(1)
/audioEncoder/blocks.2/attn/Concat_1:Concat*input:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Unsqueeze_5F_2_5F_output_5F_0(1);_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Unsqueeze_5F_3_5F_output_5F_0(1);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_5_5F_output_5F_0.bin(int64:1);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_6_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Concat_5F_1_5F_output_5F_0(4)*axis:0
/audioEncoder/blocks.2/attn/Reshape_1:Reshape*input:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_key_2F_MatMul_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Concat_5F_1_5F_output_5F_0(4)*output:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Reshape_5F_1_5F_output_5F_0()
/audioEncoder/blocks.2/attn/Transpose_1:Transpose*input:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Reshape_5F_1_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Transpose_5F_1_5F_output_5F_0()*perm:0,2,3,1
/audioEncoder/blocks.2/attn/Mul_1:Mul*input:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Transpose_5F_1_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Pow_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Mul_5F_1_5F_output_5F_0()
/audioEncoder/blocks.2/attn/Shape_5:Shape*input:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_value_2F_Add_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Shape_5F_5_5F_output_5F_0(0)
/audioEncoder/blocks.2/attn/Gather_5:Gather*input:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Shape_5F_5_5F_output_5F_0(0);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_3_5F_output_5F_0.bin(int64:)*output:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Gather_5F_5_5F_output_5F_0()*axis:0
/audioEncoder/blocks.2/attn/Gather_6:Gather*input:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Shape_5F_5_5F_output_5F_0(0);_2F_audioEncoder_2F_Constant_5F_6_5F_output_5F_0.bin(int64:)*output:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Gather_5F_6_5F_output_5F_0()*axis:0
/audioEncoder/blocks.2/attn/Unsqueeze_4:Unsqueeze*input:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Gather_5F_5_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Unsqueeze_5F_4_5F_output_5F_0(1)
/audioEncoder/blocks.2/attn/Unsqueeze_5:Unsqueeze*input:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Gather_5F_6_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Unsqueeze_5F_5_5F_output_5F_0(1)
/audioEncoder/blocks.2/attn/Concat_2:Concat*input:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Unsqueeze_5F_4_5F_output_5F_0(1);_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Unsqueeze_5F_5_5F_output_5F_0(1);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_5_5F_output_5F_0.bin(int64:1);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_6_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Concat_5F_2_5F_output_5F_0(4)*axis:0
/audioEncoder/blocks.2/attn/Reshape_2:Reshape*input:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_value_2F_Add_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Concat_5F_2_5F_output_5F_0(4)*output:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Reshape_5F_2_5F_output_5F_0()
/audioEncoder/blocks.2/attn/Transpose_2:Transpose*input:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Reshape_5F_2_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Transpose_5F_2_5F_output_5F_0()*perm:0,2,1,3
/audioEncoder/blocks.2/attn/MatMul:MatMul*input:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Mul_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Mul_5F_1_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_MatMul_5F_output_5F_0()
/audioEncoder/blocks.2/attn/Softmax:Softmax*input:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_MatMul_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Softmax_5F_output_5F_0()*axis:-1
/audioEncoder/blocks.2/attn/MatMul_1:MatMul*input:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Softmax_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Transpose_5F_2_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_MatMul_5F_1_5F_output_5F_0()
/audioEncoder/blocks.2/attn/Transpose_3:Transpose*input:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_MatMul_5F_1_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Transpose_5F_3_5F_output_5F_0()*perm:0,2,1,3
/audioEncoder/blocks.2/attn/Shape_7:Shape*input:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Transpose_5F_3_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Shape_5F_7_5F_output_5F_0(0)
/audioEncoder/blocks.2/attn/Slice:Slice*input:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Shape_5F_7_5F_output_5F_0(0);_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_17_5F_output_5F_0.bin(int64:1);_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Slice_5F_output_5F_0(0)
/audioEncoder/blocks.2/attn/Concat_3:Concat*input:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Slice_5F_output_5F_0(0);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_6_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Concat_5F_3_5F_output_5F_0(0)*axis:0
/audioEncoder/blocks.2/attn/Reshape_3:Reshape*input:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Transpose_5F_3_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Concat_5F_3_5F_output_5F_0(0)*output:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Reshape_5F_3_5F_output_5F_0()
/audioEncoder/blocks.2/attn/out/MatMul:MatMul*input:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_Reshape_5F_3_5F_output_5F_0();onnx_3A__3A_MatMul_5F_1286.bin(float32:512,512)*output:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_out_2F_MatMul_5F_output_5F_0()
/audioEncoder/blocks.2/attn/out/Add:Add*input:onnx_3A__3A_Add_5F_1285.bin(float32:512);_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_out_2F_MatMul_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_out_2F_Add_5F_output_5F_0()
/audioEncoder/blocks.2/Add:Add*input:_2F_audioEncoder_2F_blocks_2E_1_2F_Add_5F_1_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_2_2F_attn_2F_out_2F_Add_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_2_2F_Add_5F_output_5F_0()
/audioEncoder/blocks.2/mlp_ln/ReduceMean:ReduceMean*input:_2F_audioEncoder_2F_blocks_2E_2_2F_Add_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_2_2F_mlp_5F_ln_2F_ReduceMean_5F_output_5F_0()*axes:-1
/audioEncoder/blocks.2/mlp_ln/Sub:Sub*input:_2F_audioEncoder_2F_blocks_2E_2_2F_Add_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_2_2F_mlp_5F_ln_2F_ReduceMean_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_2_2F_mlp_5F_ln_2F_Sub_5F_output_5F_0()
/audioEncoder/blocks.2/mlp_ln/Pow:Pow*input:_2F_audioEncoder_2F_blocks_2E_2_2F_mlp_5F_ln_2F_Sub_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_0_2F_attn_5F_ln_2F_Constant_5F_output_5F_0.bin(float32:)*output:_2F_audioEncoder_2F_blocks_2E_2_2F_mlp_5F_ln_2F_Pow_5F_output_5F_0()
/audioEncoder/blocks.2/mlp_ln/ReduceMean_1:ReduceMean*input:_2F_audioEncoder_2F_blocks_2E_2_2F_mlp_5F_ln_2F_Pow_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_2_2F_mlp_5F_ln_2F_ReduceMean_5F_1_5F_output_5F_0()*axes:-1
/audioEncoder/blocks.2/mlp_ln/Add:Add*input:_2F_audioEncoder_2F_blocks_2E_2_2F_mlp_5F_ln_2F_ReduceMean_5F_1_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_0_2F_attn_5F_ln_2F_Constant_5F_1_5F_output_5F_0.bin(float32:)*output:_2F_audioEncoder_2F_blocks_2E_2_2F_mlp_5F_ln_2F_Add_5F_output_5F_0()
/audioEncoder/blocks.2/mlp_ln/Sqrt:Sqrt*input:_2F_audioEncoder_2F_blocks_2E_2_2F_mlp_5F_ln_2F_Add_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_2_2F_mlp_5F_ln_2F_Sqrt_5F_output_5F_0()
/audioEncoder/blocks.2/mlp_ln/Div:Div*input:_2F_audioEncoder_2F_blocks_2E_2_2F_mlp_5F_ln_2F_Sub_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_2_2F_mlp_5F_ln_2F_Sqrt_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_2_2F_mlp_5F_ln_2F_Div_5F_output_5F_0()
/audioEncoder/blocks.2/mlp_ln/Mul:Mul*input:_2F_audioEncoder_2F_blocks_2E_2_2F_mlp_5F_ln_2F_Div_5F_output_5F_0();audioEncoder_2E_blocks_2E_2_2E_mlp_5F_ln_2E_weight.bin(float32:512)*output:_2F_audioEncoder_2F_blocks_2E_2_2F_mlp_5F_ln_2F_Mul_5F_output_5F_0()
/audioEncoder/blocks.2/mlp_ln/Add_1:Add*input:_2F_audioEncoder_2F_blocks_2E_2_2F_mlp_5F_ln_2F_Mul_5F_output_5F_0();audioEncoder_2E_blocks_2E_2_2E_mlp_5F_ln_2E_bias.bin(float32:512)*output:_2F_audioEncoder_2F_blocks_2E_2_2F_mlp_5F_ln_2F_Add_5F_1_5F_output_5F_0()
/audioEncoder/blocks.2/mlp/mlp.0/MatMul:MatMul*input:_2F_audioEncoder_2F_blocks_2E_2_2F_mlp_5F_ln_2F_Add_5F_1_5F_output_5F_0();onnx_3A__3A_MatMul_5F_1289.bin(float32:512,2048)*output:_2F_audioEncoder_2F_blocks_2E_2_2F_mlp_2F_mlp_2E_0_2F_MatMul_5F_output_5F_0()
/audioEncoder/blocks.2/mlp/mlp.0/Add:Add*input:onnx_3A__3A_Add_5F_1288.bin(float32:2048);_2F_audioEncoder_2F_blocks_2E_2_2F_mlp_2F_mlp_2E_0_2F_MatMul_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_2_2F_mlp_2F_mlp_2E_0_2F_Add_5F_output_5F_0()
/audioEncoder/blocks.2/mlp/mlp.1/Div:Div*input:_2F_audioEncoder_2F_blocks_2E_2_2F_mlp_2F_mlp_2E_0_2F_Add_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_output_5F_0.bin(float32:)*output:_2F_audioEncoder_2F_blocks_2E_2_2F_mlp_2F_mlp_2E_1_2F_Div_5F_output_5F_0()
/audioEncoder/blocks.2/mlp/mlp.1/Erf:Erf*input:_2F_audioEncoder_2F_blocks_2E_2_2F_mlp_2F_mlp_2E_1_2F_Div_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_2_2F_mlp_2F_mlp_2E_1_2F_Erf_5F_output_5F_0()
/audioEncoder/blocks.2/mlp/mlp.1/Add:Add*input:_2F_audioEncoder_2F_blocks_2E_2_2F_mlp_2F_mlp_2E_1_2F_Erf_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_1_5F_output_5F_0.bin(float32:)*output:_2F_audioEncoder_2F_blocks_2E_2_2F_mlp_2F_mlp_2E_1_2F_Add_5F_output_5F_0()
/audioEncoder/blocks.2/mlp/mlp.1/Mul:Mul*input:_2F_audioEncoder_2F_blocks_2E_2_2F_mlp_2F_mlp_2E_0_2F_Add_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_2_2F_mlp_2F_mlp_2E_1_2F_Add_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_2_2F_mlp_2F_mlp_2E_1_2F_Mul_5F_output_5F_0()
/audioEncoder/blocks.2/mlp/mlp.1/Mul_1:Mul*input:_2F_audioEncoder_2F_blocks_2E_2_2F_mlp_2F_mlp_2E_1_2F_Mul_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_2_5F_output_5F_0.bin(float32:)*output:_2F_audioEncoder_2F_blocks_2E_2_2F_mlp_2F_mlp_2E_1_2F_Mul_5F_1_5F_output_5F_0()
/audioEncoder/blocks.2/mlp/mlp.2/MatMul:MatMul*input:_2F_audioEncoder_2F_blocks_2E_2_2F_mlp_2F_mlp_2E_1_2F_Mul_5F_1_5F_output_5F_0();onnx_3A__3A_MatMul_5F_1292.bin(float32:2048,512)*output:_2F_audioEncoder_2F_blocks_2E_2_2F_mlp_2F_mlp_2E_2_2F_MatMul_5F_output_5F_0()
/audioEncoder/blocks.2/mlp/mlp.2/Add:Add*input:onnx_3A__3A_Add_5F_1291.bin(float32:512);_2F_audioEncoder_2F_blocks_2E_2_2F_mlp_2F_mlp_2E_2_2F_MatMul_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_2_2F_mlp_2F_mlp_2E_2_2F_Add_5F_output_5F_0()
/audioEncoder/blocks.2/Add_1:Add*input:_2F_audioEncoder_2F_blocks_2E_2_2F_Add_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_2_2F_mlp_2F_mlp_2E_2_2F_Add_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_2_2F_Add_5F_1_5F_output_5F_0()
/audioEncoder/blocks.3/attn_ln/ReduceMean:ReduceMean*input:_2F_audioEncoder_2F_blocks_2E_2_2F_Add_5F_1_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_5F_ln_2F_ReduceMean_5F_output_5F_0()*axes:-1
/audioEncoder/blocks.3/attn_ln/Sub:Sub*input:_2F_audioEncoder_2F_blocks_2E_2_2F_Add_5F_1_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_3_2F_attn_5F_ln_2F_ReduceMean_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_5F_ln_2F_Sub_5F_output_5F_0()
/audioEncoder/blocks.3/attn_ln/Pow:Pow*input:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_5F_ln_2F_Sub_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_0_2F_attn_5F_ln_2F_Constant_5F_output_5F_0.bin(float32:)*output:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_5F_ln_2F_Pow_5F_output_5F_0()
/audioEncoder/blocks.3/attn_ln/ReduceMean_1:ReduceMean*input:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_5F_ln_2F_Pow_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_5F_ln_2F_ReduceMean_5F_1_5F_output_5F_0()*axes:-1
/audioEncoder/blocks.3/attn_ln/Add:Add*input:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_5F_ln_2F_ReduceMean_5F_1_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_0_2F_attn_5F_ln_2F_Constant_5F_1_5F_output_5F_0.bin(float32:)*output:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_5F_ln_2F_Add_5F_output_5F_0()
/audioEncoder/blocks.3/attn_ln/Sqrt:Sqrt*input:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_5F_ln_2F_Add_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_5F_ln_2F_Sqrt_5F_output_5F_0()
/audioEncoder/blocks.3/attn_ln/Div:Div*input:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_5F_ln_2F_Sub_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_3_2F_attn_5F_ln_2F_Sqrt_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_5F_ln_2F_Div_5F_output_5F_0()
/audioEncoder/blocks.3/attn_ln/Mul:Mul*input:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_5F_ln_2F_Div_5F_output_5F_0();audioEncoder_2E_blocks_2E_3_2E_attn_5F_ln_2E_weight.bin(float32:512)*output:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_5F_ln_2F_Mul_5F_output_5F_0()
/audioEncoder/blocks.3/attn_ln/Add_1:Add*input:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_5F_ln_2F_Mul_5F_output_5F_0();audioEncoder_2E_blocks_2E_3_2E_attn_5F_ln_2E_bias.bin(float32:512)*output:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_5F_ln_2F_Add_5F_1_5F_output_5F_0()
/audioEncoder/blocks.3/attn/query/MatMul:MatMul*input:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_5F_ln_2F_Add_5F_1_5F_output_5F_0();onnx_3A__3A_MatMul_5F_1295.bin(float32:512,512)*output:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_query_2F_MatMul_5F_output_5F_0()
/audioEncoder/blocks.3/attn/query/Add:Add*input:onnx_3A__3A_Add_5F_1294.bin(float32:512);_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_query_2F_MatMul_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_query_2F_Add_5F_output_5F_0()
/audioEncoder/blocks.3/attn/key/MatMul:MatMul*input:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_5F_ln_2F_Add_5F_1_5F_output_5F_0();onnx_3A__3A_MatMul_5F_1297.bin(float32:512,512)*output:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_key_2F_MatMul_5F_output_5F_0()
/audioEncoder/blocks.3/attn/value/MatMul:MatMul*input:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_5F_ln_2F_Add_5F_1_5F_output_5F_0();onnx_3A__3A_MatMul_5F_1300.bin(float32:512,512)*output:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_value_2F_MatMul_5F_output_5F_0()
/audioEncoder/blocks.3/attn/value/Add:Add*input:onnx_3A__3A_Add_5F_1299.bin(float32:512);_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_value_2F_MatMul_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_value_2F_Add_5F_output_5F_0()
/audioEncoder/blocks.3/attn/Shape:Shape*input:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_query_2F_Add_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Shape_5F_output_5F_0(0)
/audioEncoder/blocks.3/attn/Gather:Gather*input:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Shape_5F_output_5F_0(0);_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Constant_5F_output_5F_0.bin(int64:)*output:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Gather_5F_output_5F_0()*axis:0
/audioEncoder/blocks.3/attn/Div:Div*input:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Gather_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_1_5F_output_5F_0.bin(int64:)*output:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Div_5F_output_5F_0()
/audioEncoder/blocks.3/attn/Cast_2:Cast*input:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Div_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Cast_5F_2_5F_output_5F_0()*to:1
/audioEncoder/blocks.3/attn/Pow:Pow*input:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Cast_5F_2_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_2_5F_output_5F_0.bin(float32:)*output:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Pow_5F_output_5F_0()
/audioEncoder/blocks.3/attn/Gather_1:Gather*input:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Shape_5F_output_5F_0(0);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_3_5F_output_5F_0.bin(int64:)*output:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Gather_5F_1_5F_output_5F_0()*axis:0
/audioEncoder/blocks.3/attn/Gather_2:Gather*input:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Shape_5F_output_5F_0(0);_2F_audioEncoder_2F_Constant_5F_6_5F_output_5F_0.bin(int64:)*output:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Gather_5F_2_5F_output_5F_0()*axis:0
/audioEncoder/blocks.3/attn/Unsqueeze:Unsqueeze*input:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Gather_5F_1_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Unsqueeze_5F_output_5F_0(1)
/audioEncoder/blocks.3/attn/Unsqueeze_1:Unsqueeze*input:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Gather_5F_2_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Unsqueeze_5F_1_5F_output_5F_0(1)
/audioEncoder/blocks.3/attn/Concat:Concat*input:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Unsqueeze_5F_output_5F_0(1);_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Unsqueeze_5F_1_5F_output_5F_0(1);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_5_5F_output_5F_0.bin(int64:1);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_6_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Concat_5F_output_5F_0(4)*axis:0
/audioEncoder/blocks.3/attn/Reshape:Reshape*input:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_query_2F_Add_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Concat_5F_output_5F_0(4)*output:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Reshape_5F_output_5F_0()
/audioEncoder/blocks.3/attn/Transpose:Transpose*input:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Reshape_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Transpose_5F_output_5F_0()*perm:0,2,1,3
/audioEncoder/blocks.3/attn/Mul:Mul*input:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Transpose_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Pow_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Mul_5F_output_5F_0()
/audioEncoder/blocks.3/attn/Shape_3:Shape*input:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_key_2F_MatMul_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Shape_5F_3_5F_output_5F_0(0)
/audioEncoder/blocks.3/attn/Gather_3:Gather*input:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Shape_5F_3_5F_output_5F_0(0);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_3_5F_output_5F_0.bin(int64:)*output:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Gather_5F_3_5F_output_5F_0()*axis:0
/audioEncoder/blocks.3/attn/Gather_4:Gather*input:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Shape_5F_3_5F_output_5F_0(0);_2F_audioEncoder_2F_Constant_5F_6_5F_output_5F_0.bin(int64:)*output:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Gather_5F_4_5F_output_5F_0()*axis:0
/audioEncoder/blocks.3/attn/Unsqueeze_2:Unsqueeze*input:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Gather_5F_3_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Unsqueeze_5F_2_5F_output_5F_0(1)
/audioEncoder/blocks.3/attn/Unsqueeze_3:Unsqueeze*input:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Gather_5F_4_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Unsqueeze_5F_3_5F_output_5F_0(1)
/audioEncoder/blocks.3/attn/Concat_1:Concat*input:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Unsqueeze_5F_2_5F_output_5F_0(1);_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Unsqueeze_5F_3_5F_output_5F_0(1);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_5_5F_output_5F_0.bin(int64:1);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_6_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Concat_5F_1_5F_output_5F_0(4)*axis:0
/audioEncoder/blocks.3/attn/Reshape_1:Reshape*input:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_key_2F_MatMul_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Concat_5F_1_5F_output_5F_0(4)*output:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Reshape_5F_1_5F_output_5F_0()
/audioEncoder/blocks.3/attn/Transpose_1:Transpose*input:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Reshape_5F_1_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Transpose_5F_1_5F_output_5F_0()*perm:0,2,3,1
/audioEncoder/blocks.3/attn/Mul_1:Mul*input:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Transpose_5F_1_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Pow_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Mul_5F_1_5F_output_5F_0()
/audioEncoder/blocks.3/attn/Shape_5:Shape*input:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_value_2F_Add_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Shape_5F_5_5F_output_5F_0(0)
/audioEncoder/blocks.3/attn/Gather_5:Gather*input:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Shape_5F_5_5F_output_5F_0(0);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_3_5F_output_5F_0.bin(int64:)*output:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Gather_5F_5_5F_output_5F_0()*axis:0
/audioEncoder/blocks.3/attn/Gather_6:Gather*input:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Shape_5F_5_5F_output_5F_0(0);_2F_audioEncoder_2F_Constant_5F_6_5F_output_5F_0.bin(int64:)*output:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Gather_5F_6_5F_output_5F_0()*axis:0
/audioEncoder/blocks.3/attn/Unsqueeze_4:Unsqueeze*input:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Gather_5F_5_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Unsqueeze_5F_4_5F_output_5F_0(1)
/audioEncoder/blocks.3/attn/Unsqueeze_5:Unsqueeze*input:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Gather_5F_6_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Unsqueeze_5F_5_5F_output_5F_0(1)
/audioEncoder/blocks.3/attn/Concat_2:Concat*input:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Unsqueeze_5F_4_5F_output_5F_0(1);_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Unsqueeze_5F_5_5F_output_5F_0(1);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_5_5F_output_5F_0.bin(int64:1);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_6_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Concat_5F_2_5F_output_5F_0(4)*axis:0
/audioEncoder/blocks.3/attn/Reshape_2:Reshape*input:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_value_2F_Add_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Concat_5F_2_5F_output_5F_0(4)*output:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Reshape_5F_2_5F_output_5F_0()
/audioEncoder/blocks.3/attn/Transpose_2:Transpose*input:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Reshape_5F_2_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Transpose_5F_2_5F_output_5F_0()*perm:0,2,1,3
/audioEncoder/blocks.3/attn/MatMul:MatMul*input:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Mul_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Mul_5F_1_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_MatMul_5F_output_5F_0()
/audioEncoder/blocks.3/attn/Softmax:Softmax*input:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_MatMul_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Softmax_5F_output_5F_0()*axis:-1
/audioEncoder/blocks.3/attn/MatMul_1:MatMul*input:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Softmax_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Transpose_5F_2_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_MatMul_5F_1_5F_output_5F_0()
/audioEncoder/blocks.3/attn/Transpose_3:Transpose*input:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_MatMul_5F_1_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Transpose_5F_3_5F_output_5F_0()*perm:0,2,1,3
/audioEncoder/blocks.3/attn/Shape_7:Shape*input:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Transpose_5F_3_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Shape_5F_7_5F_output_5F_0(0)
/audioEncoder/blocks.3/attn/Slice:Slice*input:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Shape_5F_7_5F_output_5F_0(0);_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_17_5F_output_5F_0.bin(int64:1);_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Slice_5F_output_5F_0(0)
/audioEncoder/blocks.3/attn/Concat_3:Concat*input:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Slice_5F_output_5F_0(0);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_6_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Concat_5F_3_5F_output_5F_0(0)*axis:0
/audioEncoder/blocks.3/attn/Reshape_3:Reshape*input:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Transpose_5F_3_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Concat_5F_3_5F_output_5F_0(0)*output:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Reshape_5F_3_5F_output_5F_0()
/audioEncoder/blocks.3/attn/out/MatMul:MatMul*input:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_Reshape_5F_3_5F_output_5F_0();onnx_3A__3A_MatMul_5F_1309.bin(float32:512,512)*output:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_out_2F_MatMul_5F_output_5F_0()
/audioEncoder/blocks.3/attn/out/Add:Add*input:onnx_3A__3A_Add_5F_1308.bin(float32:512);_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_out_2F_MatMul_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_out_2F_Add_5F_output_5F_0()
/audioEncoder/blocks.3/Add:Add*input:_2F_audioEncoder_2F_blocks_2E_2_2F_Add_5F_1_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_3_2F_attn_2F_out_2F_Add_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_3_2F_Add_5F_output_5F_0()
/audioEncoder/blocks.3/mlp_ln/ReduceMean:ReduceMean*input:_2F_audioEncoder_2F_blocks_2E_3_2F_Add_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_3_2F_mlp_5F_ln_2F_ReduceMean_5F_output_5F_0()*axes:-1
/audioEncoder/blocks.3/mlp_ln/Sub:Sub*input:_2F_audioEncoder_2F_blocks_2E_3_2F_Add_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_3_2F_mlp_5F_ln_2F_ReduceMean_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_3_2F_mlp_5F_ln_2F_Sub_5F_output_5F_0()
/audioEncoder/blocks.3/mlp_ln/Pow:Pow*input:_2F_audioEncoder_2F_blocks_2E_3_2F_mlp_5F_ln_2F_Sub_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_0_2F_attn_5F_ln_2F_Constant_5F_output_5F_0.bin(float32:)*output:_2F_audioEncoder_2F_blocks_2E_3_2F_mlp_5F_ln_2F_Pow_5F_output_5F_0()
/audioEncoder/blocks.3/mlp_ln/ReduceMean_1:ReduceMean*input:_2F_audioEncoder_2F_blocks_2E_3_2F_mlp_5F_ln_2F_Pow_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_3_2F_mlp_5F_ln_2F_ReduceMean_5F_1_5F_output_5F_0()*axes:-1
/audioEncoder/blocks.3/mlp_ln/Add:Add*input:_2F_audioEncoder_2F_blocks_2E_3_2F_mlp_5F_ln_2F_ReduceMean_5F_1_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_0_2F_attn_5F_ln_2F_Constant_5F_1_5F_output_5F_0.bin(float32:)*output:_2F_audioEncoder_2F_blocks_2E_3_2F_mlp_5F_ln_2F_Add_5F_output_5F_0()
/audioEncoder/blocks.3/mlp_ln/Sqrt:Sqrt*input:_2F_audioEncoder_2F_blocks_2E_3_2F_mlp_5F_ln_2F_Add_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_3_2F_mlp_5F_ln_2F_Sqrt_5F_output_5F_0()
/audioEncoder/blocks.3/mlp_ln/Div:Div*input:_2F_audioEncoder_2F_blocks_2E_3_2F_mlp_5F_ln_2F_Sub_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_3_2F_mlp_5F_ln_2F_Sqrt_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_3_2F_mlp_5F_ln_2F_Div_5F_output_5F_0()
/audioEncoder/blocks.3/mlp_ln/Mul:Mul*input:_2F_audioEncoder_2F_blocks_2E_3_2F_mlp_5F_ln_2F_Div_5F_output_5F_0();audioEncoder_2E_blocks_2E_3_2E_mlp_5F_ln_2E_weight.bin(float32:512)*output:_2F_audioEncoder_2F_blocks_2E_3_2F_mlp_5F_ln_2F_Mul_5F_output_5F_0()
/audioEncoder/blocks.3/mlp_ln/Add_1:Add*input:_2F_audioEncoder_2F_blocks_2E_3_2F_mlp_5F_ln_2F_Mul_5F_output_5F_0();audioEncoder_2E_blocks_2E_3_2E_mlp_5F_ln_2E_bias.bin(float32:512)*output:_2F_audioEncoder_2F_blocks_2E_3_2F_mlp_5F_ln_2F_Add_5F_1_5F_output_5F_0()
/audioEncoder/blocks.3/mlp/mlp.0/MatMul:MatMul*input:_2F_audioEncoder_2F_blocks_2E_3_2F_mlp_5F_ln_2F_Add_5F_1_5F_output_5F_0();onnx_3A__3A_MatMul_5F_1312.bin(float32:512,2048)*output:_2F_audioEncoder_2F_blocks_2E_3_2F_mlp_2F_mlp_2E_0_2F_MatMul_5F_output_5F_0()
/audioEncoder/blocks.3/mlp/mlp.0/Add:Add*input:onnx_3A__3A_Add_5F_1311.bin(float32:2048);_2F_audioEncoder_2F_blocks_2E_3_2F_mlp_2F_mlp_2E_0_2F_MatMul_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_3_2F_mlp_2F_mlp_2E_0_2F_Add_5F_output_5F_0()
/audioEncoder/blocks.3/mlp/mlp.1/Div:Div*input:_2F_audioEncoder_2F_blocks_2E_3_2F_mlp_2F_mlp_2E_0_2F_Add_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_output_5F_0.bin(float32:)*output:_2F_audioEncoder_2F_blocks_2E_3_2F_mlp_2F_mlp_2E_1_2F_Div_5F_output_5F_0()
/audioEncoder/blocks.3/mlp/mlp.1/Erf:Erf*input:_2F_audioEncoder_2F_blocks_2E_3_2F_mlp_2F_mlp_2E_1_2F_Div_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_3_2F_mlp_2F_mlp_2E_1_2F_Erf_5F_output_5F_0()
/audioEncoder/blocks.3/mlp/mlp.1/Add:Add*input:_2F_audioEncoder_2F_blocks_2E_3_2F_mlp_2F_mlp_2E_1_2F_Erf_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_1_5F_output_5F_0.bin(float32:)*output:_2F_audioEncoder_2F_blocks_2E_3_2F_mlp_2F_mlp_2E_1_2F_Add_5F_output_5F_0()
/audioEncoder/blocks.3/mlp/mlp.1/Mul:Mul*input:_2F_audioEncoder_2F_blocks_2E_3_2F_mlp_2F_mlp_2E_0_2F_Add_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_3_2F_mlp_2F_mlp_2E_1_2F_Add_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_3_2F_mlp_2F_mlp_2E_1_2F_Mul_5F_output_5F_0()
/audioEncoder/blocks.3/mlp/mlp.1/Mul_1:Mul*input:_2F_audioEncoder_2F_blocks_2E_3_2F_mlp_2F_mlp_2E_1_2F_Mul_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_2_5F_output_5F_0.bin(float32:)*output:_2F_audioEncoder_2F_blocks_2E_3_2F_mlp_2F_mlp_2E_1_2F_Mul_5F_1_5F_output_5F_0()
/audioEncoder/blocks.3/mlp/mlp.2/MatMul:MatMul*input:_2F_audioEncoder_2F_blocks_2E_3_2F_mlp_2F_mlp_2E_1_2F_Mul_5F_1_5F_output_5F_0();onnx_3A__3A_MatMul_5F_1315.bin(float32:2048,512)*output:_2F_audioEncoder_2F_blocks_2E_3_2F_mlp_2F_mlp_2E_2_2F_MatMul_5F_output_5F_0()
/audioEncoder/blocks.3/mlp/mlp.2/Add:Add*input:onnx_3A__3A_Add_5F_1314.bin(float32:512);_2F_audioEncoder_2F_blocks_2E_3_2F_mlp_2F_mlp_2E_2_2F_MatMul_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_3_2F_mlp_2F_mlp_2E_2_2F_Add_5F_output_5F_0()
/audioEncoder/blocks.3/Add_1:Add*input:_2F_audioEncoder_2F_blocks_2E_3_2F_Add_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_3_2F_mlp_2F_mlp_2E_2_2F_Add_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_3_2F_Add_5F_1_5F_output_5F_0()
/audioEncoder/blocks.4/attn_ln/ReduceMean:ReduceMean*input:_2F_audioEncoder_2F_blocks_2E_3_2F_Add_5F_1_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_5F_ln_2F_ReduceMean_5F_output_5F_0()*axes:-1
/audioEncoder/blocks.4/attn_ln/Sub:Sub*input:_2F_audioEncoder_2F_blocks_2E_3_2F_Add_5F_1_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_4_2F_attn_5F_ln_2F_ReduceMean_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_5F_ln_2F_Sub_5F_output_5F_0()
/audioEncoder/blocks.4/attn_ln/Pow:Pow*input:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_5F_ln_2F_Sub_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_0_2F_attn_5F_ln_2F_Constant_5F_output_5F_0.bin(float32:)*output:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_5F_ln_2F_Pow_5F_output_5F_0()
/audioEncoder/blocks.4/attn_ln/ReduceMean_1:ReduceMean*input:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_5F_ln_2F_Pow_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_5F_ln_2F_ReduceMean_5F_1_5F_output_5F_0()*axes:-1
/audioEncoder/blocks.4/attn_ln/Add:Add*input:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_5F_ln_2F_ReduceMean_5F_1_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_0_2F_attn_5F_ln_2F_Constant_5F_1_5F_output_5F_0.bin(float32:)*output:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_5F_ln_2F_Add_5F_output_5F_0()
/audioEncoder/blocks.4/attn_ln/Sqrt:Sqrt*input:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_5F_ln_2F_Add_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_5F_ln_2F_Sqrt_5F_output_5F_0()
/audioEncoder/blocks.4/attn_ln/Div:Div*input:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_5F_ln_2F_Sub_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_4_2F_attn_5F_ln_2F_Sqrt_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_5F_ln_2F_Div_5F_output_5F_0()
/audioEncoder/blocks.4/attn_ln/Mul:Mul*input:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_5F_ln_2F_Div_5F_output_5F_0();audioEncoder_2E_blocks_2E_4_2E_attn_5F_ln_2E_weight.bin(float32:512)*output:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_5F_ln_2F_Mul_5F_output_5F_0()
/audioEncoder/blocks.4/attn_ln/Add_1:Add*input:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_5F_ln_2F_Mul_5F_output_5F_0();audioEncoder_2E_blocks_2E_4_2E_attn_5F_ln_2E_bias.bin(float32:512)*output:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_5F_ln_2F_Add_5F_1_5F_output_5F_0()
/audioEncoder/blocks.4/attn/query/MatMul:MatMul*input:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_5F_ln_2F_Add_5F_1_5F_output_5F_0();onnx_3A__3A_MatMul_5F_1318.bin(float32:512,512)*output:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_query_2F_MatMul_5F_output_5F_0()
/audioEncoder/blocks.4/attn/query/Add:Add*input:onnx_3A__3A_Add_5F_1317.bin(float32:512);_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_query_2F_MatMul_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_query_2F_Add_5F_output_5F_0()
/audioEncoder/blocks.4/attn/key/MatMul:MatMul*input:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_5F_ln_2F_Add_5F_1_5F_output_5F_0();onnx_3A__3A_MatMul_5F_1320.bin(float32:512,512)*output:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_key_2F_MatMul_5F_output_5F_0()
/audioEncoder/blocks.4/attn/value/MatMul:MatMul*input:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_5F_ln_2F_Add_5F_1_5F_output_5F_0();onnx_3A__3A_MatMul_5F_1323.bin(float32:512,512)*output:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_value_2F_MatMul_5F_output_5F_0()
/audioEncoder/blocks.4/attn/value/Add:Add*input:onnx_3A__3A_Add_5F_1322.bin(float32:512);_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_value_2F_MatMul_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_value_2F_Add_5F_output_5F_0()
/audioEncoder/blocks.4/attn/Shape:Shape*input:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_query_2F_Add_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Shape_5F_output_5F_0(0)
/audioEncoder/blocks.4/attn/Gather:Gather*input:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Shape_5F_output_5F_0(0);_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Constant_5F_output_5F_0.bin(int64:)*output:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Gather_5F_output_5F_0()*axis:0
/audioEncoder/blocks.4/attn/Div:Div*input:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Gather_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_1_5F_output_5F_0.bin(int64:)*output:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Div_5F_output_5F_0()
/audioEncoder/blocks.4/attn/Cast_2:Cast*input:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Div_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Cast_5F_2_5F_output_5F_0()*to:1
/audioEncoder/blocks.4/attn/Pow:Pow*input:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Cast_5F_2_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_2_5F_output_5F_0.bin(float32:)*output:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Pow_5F_output_5F_0()
/audioEncoder/blocks.4/attn/Gather_1:Gather*input:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Shape_5F_output_5F_0(0);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_3_5F_output_5F_0.bin(int64:)*output:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Gather_5F_1_5F_output_5F_0()*axis:0
/audioEncoder/blocks.4/attn/Gather_2:Gather*input:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Shape_5F_output_5F_0(0);_2F_audioEncoder_2F_Constant_5F_6_5F_output_5F_0.bin(int64:)*output:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Gather_5F_2_5F_output_5F_0()*axis:0
/audioEncoder/blocks.4/attn/Unsqueeze:Unsqueeze*input:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Gather_5F_1_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Unsqueeze_5F_output_5F_0(1)
/audioEncoder/blocks.4/attn/Unsqueeze_1:Unsqueeze*input:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Gather_5F_2_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Unsqueeze_5F_1_5F_output_5F_0(1)
/audioEncoder/blocks.4/attn/Concat:Concat*input:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Unsqueeze_5F_output_5F_0(1);_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Unsqueeze_5F_1_5F_output_5F_0(1);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_5_5F_output_5F_0.bin(int64:1);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_6_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Concat_5F_output_5F_0(4)*axis:0
/audioEncoder/blocks.4/attn/Reshape:Reshape*input:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_query_2F_Add_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Concat_5F_output_5F_0(4)*output:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Reshape_5F_output_5F_0()
/audioEncoder/blocks.4/attn/Transpose:Transpose*input:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Reshape_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Transpose_5F_output_5F_0()*perm:0,2,1,3
/audioEncoder/blocks.4/attn/Mul:Mul*input:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Transpose_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Pow_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Mul_5F_output_5F_0()
/audioEncoder/blocks.4/attn/Shape_3:Shape*input:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_key_2F_MatMul_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Shape_5F_3_5F_output_5F_0(0)
/audioEncoder/blocks.4/attn/Gather_3:Gather*input:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Shape_5F_3_5F_output_5F_0(0);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_3_5F_output_5F_0.bin(int64:)*output:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Gather_5F_3_5F_output_5F_0()*axis:0
/audioEncoder/blocks.4/attn/Gather_4:Gather*input:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Shape_5F_3_5F_output_5F_0(0);_2F_audioEncoder_2F_Constant_5F_6_5F_output_5F_0.bin(int64:)*output:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Gather_5F_4_5F_output_5F_0()*axis:0
/audioEncoder/blocks.4/attn/Unsqueeze_2:Unsqueeze*input:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Gather_5F_3_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Unsqueeze_5F_2_5F_output_5F_0(1)
/audioEncoder/blocks.4/attn/Unsqueeze_3:Unsqueeze*input:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Gather_5F_4_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Unsqueeze_5F_3_5F_output_5F_0(1)
/audioEncoder/blocks.4/attn/Concat_1:Concat*input:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Unsqueeze_5F_2_5F_output_5F_0(1);_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Unsqueeze_5F_3_5F_output_5F_0(1);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_5_5F_output_5F_0.bin(int64:1);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_6_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Concat_5F_1_5F_output_5F_0(4)*axis:0
/audioEncoder/blocks.4/attn/Reshape_1:Reshape*input:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_key_2F_MatMul_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Concat_5F_1_5F_output_5F_0(4)*output:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Reshape_5F_1_5F_output_5F_0()
/audioEncoder/blocks.4/attn/Transpose_1:Transpose*input:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Reshape_5F_1_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Transpose_5F_1_5F_output_5F_0()*perm:0,2,3,1
/audioEncoder/blocks.4/attn/Mul_1:Mul*input:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Transpose_5F_1_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Pow_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Mul_5F_1_5F_output_5F_0()
/audioEncoder/blocks.4/attn/Shape_5:Shape*input:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_value_2F_Add_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Shape_5F_5_5F_output_5F_0(0)
/audioEncoder/blocks.4/attn/Gather_5:Gather*input:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Shape_5F_5_5F_output_5F_0(0);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_3_5F_output_5F_0.bin(int64:)*output:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Gather_5F_5_5F_output_5F_0()*axis:0
/audioEncoder/blocks.4/attn/Gather_6:Gather*input:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Shape_5F_5_5F_output_5F_0(0);_2F_audioEncoder_2F_Constant_5F_6_5F_output_5F_0.bin(int64:)*output:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Gather_5F_6_5F_output_5F_0()*axis:0
/audioEncoder/blocks.4/attn/Unsqueeze_4:Unsqueeze*input:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Gather_5F_5_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Unsqueeze_5F_4_5F_output_5F_0(1)
/audioEncoder/blocks.4/attn/Unsqueeze_5:Unsqueeze*input:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Gather_5F_6_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Unsqueeze_5F_5_5F_output_5F_0(1)
/audioEncoder/blocks.4/attn/Concat_2:Concat*input:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Unsqueeze_5F_4_5F_output_5F_0(1);_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Unsqueeze_5F_5_5F_output_5F_0(1);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_5_5F_output_5F_0.bin(int64:1);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_6_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Concat_5F_2_5F_output_5F_0(4)*axis:0
/audioEncoder/blocks.4/attn/Reshape_2:Reshape*input:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_value_2F_Add_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Concat_5F_2_5F_output_5F_0(4)*output:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Reshape_5F_2_5F_output_5F_0()
/audioEncoder/blocks.4/attn/Transpose_2:Transpose*input:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Reshape_5F_2_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Transpose_5F_2_5F_output_5F_0()*perm:0,2,1,3
/audioEncoder/blocks.4/attn/MatMul:MatMul*input:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Mul_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Mul_5F_1_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_MatMul_5F_output_5F_0()
/audioEncoder/blocks.4/attn/Softmax:Softmax*input:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_MatMul_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Softmax_5F_output_5F_0()*axis:-1
/audioEncoder/blocks.4/attn/MatMul_1:MatMul*input:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Softmax_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Transpose_5F_2_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_MatMul_5F_1_5F_output_5F_0()
/audioEncoder/blocks.4/attn/Transpose_3:Transpose*input:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_MatMul_5F_1_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Transpose_5F_3_5F_output_5F_0()*perm:0,2,1,3
/audioEncoder/blocks.4/attn/Shape_7:Shape*input:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Transpose_5F_3_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Shape_5F_7_5F_output_5F_0(0)
/audioEncoder/blocks.4/attn/Slice:Slice*input:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Shape_5F_7_5F_output_5F_0(0);_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_17_5F_output_5F_0.bin(int64:1);_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Slice_5F_output_5F_0(0)
/audioEncoder/blocks.4/attn/Concat_3:Concat*input:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Slice_5F_output_5F_0(0);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_6_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Concat_5F_3_5F_output_5F_0(0)*axis:0
/audioEncoder/blocks.4/attn/Reshape_3:Reshape*input:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Transpose_5F_3_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Concat_5F_3_5F_output_5F_0(0)*output:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Reshape_5F_3_5F_output_5F_0()
/audioEncoder/blocks.4/attn/out/MatMul:MatMul*input:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_Reshape_5F_3_5F_output_5F_0();onnx_3A__3A_MatMul_5F_1332.bin(float32:512,512)*output:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_out_2F_MatMul_5F_output_5F_0()
/audioEncoder/blocks.4/attn/out/Add:Add*input:onnx_3A__3A_Add_5F_1331.bin(float32:512);_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_out_2F_MatMul_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_out_2F_Add_5F_output_5F_0()
/audioEncoder/blocks.4/Add:Add*input:_2F_audioEncoder_2F_blocks_2E_3_2F_Add_5F_1_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_4_2F_attn_2F_out_2F_Add_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_4_2F_Add_5F_output_5F_0()
/audioEncoder/blocks.4/mlp_ln/ReduceMean:ReduceMean*input:_2F_audioEncoder_2F_blocks_2E_4_2F_Add_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_4_2F_mlp_5F_ln_2F_ReduceMean_5F_output_5F_0()*axes:-1
/audioEncoder/blocks.4/mlp_ln/Sub:Sub*input:_2F_audioEncoder_2F_blocks_2E_4_2F_Add_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_4_2F_mlp_5F_ln_2F_ReduceMean_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_4_2F_mlp_5F_ln_2F_Sub_5F_output_5F_0()
/audioEncoder/blocks.4/mlp_ln/Pow:Pow*input:_2F_audioEncoder_2F_blocks_2E_4_2F_mlp_5F_ln_2F_Sub_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_0_2F_attn_5F_ln_2F_Constant_5F_output_5F_0.bin(float32:)*output:_2F_audioEncoder_2F_blocks_2E_4_2F_mlp_5F_ln_2F_Pow_5F_output_5F_0()
/audioEncoder/blocks.4/mlp_ln/ReduceMean_1:ReduceMean*input:_2F_audioEncoder_2F_blocks_2E_4_2F_mlp_5F_ln_2F_Pow_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_4_2F_mlp_5F_ln_2F_ReduceMean_5F_1_5F_output_5F_0()*axes:-1
/audioEncoder/blocks.4/mlp_ln/Add:Add*input:_2F_audioEncoder_2F_blocks_2E_4_2F_mlp_5F_ln_2F_ReduceMean_5F_1_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_0_2F_attn_5F_ln_2F_Constant_5F_1_5F_output_5F_0.bin(float32:)*output:_2F_audioEncoder_2F_blocks_2E_4_2F_mlp_5F_ln_2F_Add_5F_output_5F_0()
/audioEncoder/blocks.4/mlp_ln/Sqrt:Sqrt*input:_2F_audioEncoder_2F_blocks_2E_4_2F_mlp_5F_ln_2F_Add_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_4_2F_mlp_5F_ln_2F_Sqrt_5F_output_5F_0()
/audioEncoder/blocks.4/mlp_ln/Div:Div*input:_2F_audioEncoder_2F_blocks_2E_4_2F_mlp_5F_ln_2F_Sub_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_4_2F_mlp_5F_ln_2F_Sqrt_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_4_2F_mlp_5F_ln_2F_Div_5F_output_5F_0()
/audioEncoder/blocks.4/mlp_ln/Mul:Mul*input:_2F_audioEncoder_2F_blocks_2E_4_2F_mlp_5F_ln_2F_Div_5F_output_5F_0();audioEncoder_2E_blocks_2E_4_2E_mlp_5F_ln_2E_weight.bin(float32:512)*output:_2F_audioEncoder_2F_blocks_2E_4_2F_mlp_5F_ln_2F_Mul_5F_output_5F_0()
/audioEncoder/blocks.4/mlp_ln/Add_1:Add*input:_2F_audioEncoder_2F_blocks_2E_4_2F_mlp_5F_ln_2F_Mul_5F_output_5F_0();audioEncoder_2E_blocks_2E_4_2E_mlp_5F_ln_2E_bias.bin(float32:512)*output:_2F_audioEncoder_2F_blocks_2E_4_2F_mlp_5F_ln_2F_Add_5F_1_5F_output_5F_0()
/audioEncoder/blocks.4/mlp/mlp.0/MatMul:MatMul*input:_2F_audioEncoder_2F_blocks_2E_4_2F_mlp_5F_ln_2F_Add_5F_1_5F_output_5F_0();onnx_3A__3A_MatMul_5F_1335.bin(float32:512,2048)*output:_2F_audioEncoder_2F_blocks_2E_4_2F_mlp_2F_mlp_2E_0_2F_MatMul_5F_output_5F_0()
/audioEncoder/blocks.4/mlp/mlp.0/Add:Add*input:onnx_3A__3A_Add_5F_1334.bin(float32:2048);_2F_audioEncoder_2F_blocks_2E_4_2F_mlp_2F_mlp_2E_0_2F_MatMul_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_4_2F_mlp_2F_mlp_2E_0_2F_Add_5F_output_5F_0()
/audioEncoder/blocks.4/mlp/mlp.1/Div:Div*input:_2F_audioEncoder_2F_blocks_2E_4_2F_mlp_2F_mlp_2E_0_2F_Add_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_output_5F_0.bin(float32:)*output:_2F_audioEncoder_2F_blocks_2E_4_2F_mlp_2F_mlp_2E_1_2F_Div_5F_output_5F_0()
/audioEncoder/blocks.4/mlp/mlp.1/Erf:Erf*input:_2F_audioEncoder_2F_blocks_2E_4_2F_mlp_2F_mlp_2E_1_2F_Div_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_4_2F_mlp_2F_mlp_2E_1_2F_Erf_5F_output_5F_0()
/audioEncoder/blocks.4/mlp/mlp.1/Add:Add*input:_2F_audioEncoder_2F_blocks_2E_4_2F_mlp_2F_mlp_2E_1_2F_Erf_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_1_5F_output_5F_0.bin(float32:)*output:_2F_audioEncoder_2F_blocks_2E_4_2F_mlp_2F_mlp_2E_1_2F_Add_5F_output_5F_0()
/audioEncoder/blocks.4/mlp/mlp.1/Mul:Mul*input:_2F_audioEncoder_2F_blocks_2E_4_2F_mlp_2F_mlp_2E_0_2F_Add_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_4_2F_mlp_2F_mlp_2E_1_2F_Add_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_4_2F_mlp_2F_mlp_2E_1_2F_Mul_5F_output_5F_0()
/audioEncoder/blocks.4/mlp/mlp.1/Mul_1:Mul*input:_2F_audioEncoder_2F_blocks_2E_4_2F_mlp_2F_mlp_2E_1_2F_Mul_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_2_5F_output_5F_0.bin(float32:)*output:_2F_audioEncoder_2F_blocks_2E_4_2F_mlp_2F_mlp_2E_1_2F_Mul_5F_1_5F_output_5F_0()
/audioEncoder/blocks.4/mlp/mlp.2/MatMul:MatMul*input:_2F_audioEncoder_2F_blocks_2E_4_2F_mlp_2F_mlp_2E_1_2F_Mul_5F_1_5F_output_5F_0();onnx_3A__3A_MatMul_5F_1338.bin(float32:2048,512)*output:_2F_audioEncoder_2F_blocks_2E_4_2F_mlp_2F_mlp_2E_2_2F_MatMul_5F_output_5F_0()
/audioEncoder/blocks.4/mlp/mlp.2/Add:Add*input:onnx_3A__3A_Add_5F_1337.bin(float32:512);_2F_audioEncoder_2F_blocks_2E_4_2F_mlp_2F_mlp_2E_2_2F_MatMul_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_4_2F_mlp_2F_mlp_2E_2_2F_Add_5F_output_5F_0()
/audioEncoder/blocks.4/Add_1:Add*input:_2F_audioEncoder_2F_blocks_2E_4_2F_Add_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_4_2F_mlp_2F_mlp_2E_2_2F_Add_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_4_2F_Add_5F_1_5F_output_5F_0()
/audioEncoder/blocks.5/attn_ln/ReduceMean:ReduceMean*input:_2F_audioEncoder_2F_blocks_2E_4_2F_Add_5F_1_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_5F_ln_2F_ReduceMean_5F_output_5F_0()*axes:-1
/audioEncoder/blocks.5/attn_ln/Sub:Sub*input:_2F_audioEncoder_2F_blocks_2E_4_2F_Add_5F_1_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_5_2F_attn_5F_ln_2F_ReduceMean_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_5F_ln_2F_Sub_5F_output_5F_0()
/audioEncoder/blocks.5/attn_ln/Pow:Pow*input:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_5F_ln_2F_Sub_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_0_2F_attn_5F_ln_2F_Constant_5F_output_5F_0.bin(float32:)*output:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_5F_ln_2F_Pow_5F_output_5F_0()
/audioEncoder/blocks.5/attn_ln/ReduceMean_1:ReduceMean*input:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_5F_ln_2F_Pow_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_5F_ln_2F_ReduceMean_5F_1_5F_output_5F_0()*axes:-1
/audioEncoder/blocks.5/attn_ln/Add:Add*input:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_5F_ln_2F_ReduceMean_5F_1_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_0_2F_attn_5F_ln_2F_Constant_5F_1_5F_output_5F_0.bin(float32:)*output:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_5F_ln_2F_Add_5F_output_5F_0()
/audioEncoder/blocks.5/attn_ln/Sqrt:Sqrt*input:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_5F_ln_2F_Add_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_5F_ln_2F_Sqrt_5F_output_5F_0()
/audioEncoder/blocks.5/attn_ln/Div:Div*input:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_5F_ln_2F_Sub_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_5_2F_attn_5F_ln_2F_Sqrt_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_5F_ln_2F_Div_5F_output_5F_0()
/audioEncoder/blocks.5/attn_ln/Mul:Mul*input:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_5F_ln_2F_Div_5F_output_5F_0();audioEncoder_2E_blocks_2E_5_2E_attn_5F_ln_2E_weight.bin(float32:512)*output:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_5F_ln_2F_Mul_5F_output_5F_0()
/audioEncoder/blocks.5/attn_ln/Add_1:Add*input:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_5F_ln_2F_Mul_5F_output_5F_0();audioEncoder_2E_blocks_2E_5_2E_attn_5F_ln_2E_bias.bin(float32:512)*output:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_5F_ln_2F_Add_5F_1_5F_output_5F_0()
/audioEncoder/blocks.5/attn/query/MatMul:MatMul*input:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_5F_ln_2F_Add_5F_1_5F_output_5F_0();onnx_3A__3A_MatMul_5F_1341.bin(float32:512,512)*output:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_query_2F_MatMul_5F_output_5F_0()
/audioEncoder/blocks.5/attn/query/Add:Add*input:onnx_3A__3A_Add_5F_1340.bin(float32:512);_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_query_2F_MatMul_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_query_2F_Add_5F_output_5F_0()
/audioEncoder/blocks.5/attn/key/MatMul:MatMul*input:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_5F_ln_2F_Add_5F_1_5F_output_5F_0();onnx_3A__3A_MatMul_5F_1343.bin(float32:512,512)*output:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_key_2F_MatMul_5F_output_5F_0()
/audioEncoder/blocks.5/attn/value/MatMul:MatMul*input:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_5F_ln_2F_Add_5F_1_5F_output_5F_0();onnx_3A__3A_MatMul_5F_1346.bin(float32:512,512)*output:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_value_2F_MatMul_5F_output_5F_0()
/audioEncoder/blocks.5/attn/value/Add:Add*input:onnx_3A__3A_Add_5F_1345.bin(float32:512);_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_value_2F_MatMul_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_value_2F_Add_5F_output_5F_0()
/audioEncoder/blocks.5/attn/Shape:Shape*input:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_query_2F_Add_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Shape_5F_output_5F_0(0)
/audioEncoder/blocks.5/attn/Gather:Gather*input:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Shape_5F_output_5F_0(0);_2F_audioEncoder_2F_blocks_2E_1_2F_attn_2F_Constant_5F_output_5F_0.bin(int64:)*output:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Gather_5F_output_5F_0()*axis:0
/audioEncoder/blocks.5/attn/Div:Div*input:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Gather_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_1_5F_output_5F_0.bin(int64:)*output:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Div_5F_output_5F_0()
/audioEncoder/blocks.5/attn/Cast_2:Cast*input:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Div_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Cast_5F_2_5F_output_5F_0()*to:1
/audioEncoder/blocks.5/attn/Pow:Pow*input:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Cast_5F_2_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_2_5F_output_5F_0.bin(float32:)*output:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Pow_5F_output_5F_0()
/audioEncoder/blocks.5/attn/Gather_1:Gather*input:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Shape_5F_output_5F_0(0);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_3_5F_output_5F_0.bin(int64:)*output:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Gather_5F_1_5F_output_5F_0()*axis:0
/audioEncoder/blocks.5/attn/Gather_2:Gather*input:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Shape_5F_output_5F_0(0);_2F_audioEncoder_2F_Constant_5F_6_5F_output_5F_0.bin(int64:)*output:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Gather_5F_2_5F_output_5F_0()*axis:0
/audioEncoder/blocks.5/attn/Unsqueeze:Unsqueeze*input:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Gather_5F_1_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Unsqueeze_5F_output_5F_0(1)
/audioEncoder/blocks.5/attn/Unsqueeze_1:Unsqueeze*input:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Gather_5F_2_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Unsqueeze_5F_1_5F_output_5F_0(1)
/audioEncoder/blocks.5/attn/Concat:Concat*input:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Unsqueeze_5F_output_5F_0(1);_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Unsqueeze_5F_1_5F_output_5F_0(1);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_5_5F_output_5F_0.bin(int64:1);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_6_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Concat_5F_output_5F_0(4)*axis:0
/audioEncoder/blocks.5/attn/Reshape:Reshape*input:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_query_2F_Add_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Concat_5F_output_5F_0(4)*output:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Reshape_5F_output_5F_0()
/audioEncoder/blocks.5/attn/Transpose:Transpose*input:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Reshape_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Transpose_5F_output_5F_0()*perm:0,2,1,3
/audioEncoder/blocks.5/attn/Mul:Mul*input:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Transpose_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Pow_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Mul_5F_output_5F_0()
/audioEncoder/blocks.5/attn/Shape_3:Shape*input:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_key_2F_MatMul_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Shape_5F_3_5F_output_5F_0(0)
/audioEncoder/blocks.5/attn/Gather_3:Gather*input:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Shape_5F_3_5F_output_5F_0(0);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_3_5F_output_5F_0.bin(int64:)*output:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Gather_5F_3_5F_output_5F_0()*axis:0
/audioEncoder/blocks.5/attn/Gather_4:Gather*input:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Shape_5F_3_5F_output_5F_0(0);_2F_audioEncoder_2F_Constant_5F_6_5F_output_5F_0.bin(int64:)*output:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Gather_5F_4_5F_output_5F_0()*axis:0
/audioEncoder/blocks.5/attn/Unsqueeze_2:Unsqueeze*input:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Gather_5F_3_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Unsqueeze_5F_2_5F_output_5F_0(1)
/audioEncoder/blocks.5/attn/Unsqueeze_3:Unsqueeze*input:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Gather_5F_4_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Unsqueeze_5F_3_5F_output_5F_0(1)
/audioEncoder/blocks.5/attn/Concat_1:Concat*input:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Unsqueeze_5F_2_5F_output_5F_0(1);_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Unsqueeze_5F_3_5F_output_5F_0(1);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_5_5F_output_5F_0.bin(int64:1);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_6_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Concat_5F_1_5F_output_5F_0(4)*axis:0
/audioEncoder/blocks.5/attn/Reshape_1:Reshape*input:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_key_2F_MatMul_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Concat_5F_1_5F_output_5F_0(4)*output:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Reshape_5F_1_5F_output_5F_0()
/audioEncoder/blocks.5/attn/Transpose_1:Transpose*input:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Reshape_5F_1_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Transpose_5F_1_5F_output_5F_0()*perm:0,2,3,1
/audioEncoder/blocks.5/attn/Mul_1:Mul*input:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Transpose_5F_1_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Pow_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Mul_5F_1_5F_output_5F_0()
/audioEncoder/blocks.5/attn/Shape_5:Shape*input:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_value_2F_Add_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Shape_5F_5_5F_output_5F_0(0)
/audioEncoder/blocks.5/attn/Gather_5:Gather*input:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Shape_5F_5_5F_output_5F_0(0);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_3_5F_output_5F_0.bin(int64:)*output:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Gather_5F_5_5F_output_5F_0()*axis:0
/audioEncoder/blocks.5/attn/Gather_6:Gather*input:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Shape_5F_5_5F_output_5F_0(0);_2F_audioEncoder_2F_Constant_5F_6_5F_output_5F_0.bin(int64:)*output:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Gather_5F_6_5F_output_5F_0()*axis:0
/audioEncoder/blocks.5/attn/Unsqueeze_4:Unsqueeze*input:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Gather_5F_5_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Unsqueeze_5F_4_5F_output_5F_0(1)
/audioEncoder/blocks.5/attn/Unsqueeze_5:Unsqueeze*input:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Gather_5F_6_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Unsqueeze_5F_5_5F_output_5F_0(1)
/audioEncoder/blocks.5/attn/Concat_2:Concat*input:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Unsqueeze_5F_4_5F_output_5F_0(1);_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Unsqueeze_5F_5_5F_output_5F_0(1);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_5_5F_output_5F_0.bin(int64:1);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_6_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Concat_5F_2_5F_output_5F_0(4)*axis:0
/audioEncoder/blocks.5/attn/Reshape_2:Reshape*input:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_value_2F_Add_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Concat_5F_2_5F_output_5F_0(4)*output:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Reshape_5F_2_5F_output_5F_0()
/audioEncoder/blocks.5/attn/Transpose_2:Transpose*input:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Reshape_5F_2_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Transpose_5F_2_5F_output_5F_0()*perm:0,2,1,3
/audioEncoder/blocks.5/attn/MatMul:MatMul*input:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Mul_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Mul_5F_1_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_MatMul_5F_output_5F_0()
/audioEncoder/blocks.5/attn/Softmax:Softmax*input:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_MatMul_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Softmax_5F_output_5F_0()*axis:-1
/audioEncoder/blocks.5/attn/MatMul_1:MatMul*input:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Softmax_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Transpose_5F_2_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_MatMul_5F_1_5F_output_5F_0()
/audioEncoder/blocks.5/attn/Transpose_3:Transpose*input:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_MatMul_5F_1_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Transpose_5F_3_5F_output_5F_0()*perm:0,2,1,3
/audioEncoder/blocks.5/attn/Shape_7:Shape*input:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Transpose_5F_3_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Shape_5F_7_5F_output_5F_0(0)
/audioEncoder/blocks.5/attn/Slice:Slice*input:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Shape_5F_7_5F_output_5F_0(0);_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_17_5F_output_5F_0.bin(int64:1);_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Slice_5F_output_5F_0(0)
/audioEncoder/blocks.5/attn/Concat_3:Concat*input:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Slice_5F_output_5F_0(0);_2F_audioEncoder_2F_blocks_2E_0_2F_attn_2F_Constant_5F_6_5F_output_5F_0.bin(int64:1)*output:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Concat_5F_3_5F_output_5F_0(0)*axis:0
/audioEncoder/blocks.5/attn/Reshape_3:Reshape*input:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Transpose_5F_3_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Concat_5F_3_5F_output_5F_0(0)*output:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Reshape_5F_3_5F_output_5F_0()
/audioEncoder/blocks.5/attn/out/MatMul:MatMul*input:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_Reshape_5F_3_5F_output_5F_0();onnx_3A__3A_MatMul_5F_1355.bin(float32:512,512)*output:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_out_2F_MatMul_5F_output_5F_0()
/audioEncoder/blocks.5/attn/out/Add:Add*input:onnx_3A__3A_Add_5F_1354.bin(float32:512);_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_out_2F_MatMul_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_out_2F_Add_5F_output_5F_0()
/audioEncoder/blocks.5/Add:Add*input:_2F_audioEncoder_2F_blocks_2E_4_2F_Add_5F_1_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_5_2F_attn_2F_out_2F_Add_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_5_2F_Add_5F_output_5F_0()
/audioEncoder/blocks.5/mlp_ln/ReduceMean:ReduceMean*input:_2F_audioEncoder_2F_blocks_2E_5_2F_Add_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_5_2F_mlp_5F_ln_2F_ReduceMean_5F_output_5F_0()*axes:-1
/audioEncoder/blocks.5/mlp_ln/Sub:Sub*input:_2F_audioEncoder_2F_blocks_2E_5_2F_Add_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_5_2F_mlp_5F_ln_2F_ReduceMean_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_5_2F_mlp_5F_ln_2F_Sub_5F_output_5F_0()
/audioEncoder/blocks.5/mlp_ln/Pow:Pow*input:_2F_audioEncoder_2F_blocks_2E_5_2F_mlp_5F_ln_2F_Sub_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_0_2F_attn_5F_ln_2F_Constant_5F_output_5F_0.bin(float32:)*output:_2F_audioEncoder_2F_blocks_2E_5_2F_mlp_5F_ln_2F_Pow_5F_output_5F_0()
/audioEncoder/blocks.5/mlp_ln/ReduceMean_1:ReduceMean*input:_2F_audioEncoder_2F_blocks_2E_5_2F_mlp_5F_ln_2F_Pow_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_5_2F_mlp_5F_ln_2F_ReduceMean_5F_1_5F_output_5F_0()*axes:-1
/audioEncoder/blocks.5/mlp_ln/Add:Add*input:_2F_audioEncoder_2F_blocks_2E_5_2F_mlp_5F_ln_2F_ReduceMean_5F_1_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_0_2F_attn_5F_ln_2F_Constant_5F_1_5F_output_5F_0.bin(float32:)*output:_2F_audioEncoder_2F_blocks_2E_5_2F_mlp_5F_ln_2F_Add_5F_output_5F_0()
/audioEncoder/blocks.5/mlp_ln/Sqrt:Sqrt*input:_2F_audioEncoder_2F_blocks_2E_5_2F_mlp_5F_ln_2F_Add_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_5_2F_mlp_5F_ln_2F_Sqrt_5F_output_5F_0()
/audioEncoder/blocks.5/mlp_ln/Div:Div*input:_2F_audioEncoder_2F_blocks_2E_5_2F_mlp_5F_ln_2F_Sub_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_5_2F_mlp_5F_ln_2F_Sqrt_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_5_2F_mlp_5F_ln_2F_Div_5F_output_5F_0()
/audioEncoder/blocks.5/mlp_ln/Mul:Mul*input:_2F_audioEncoder_2F_blocks_2E_5_2F_mlp_5F_ln_2F_Div_5F_output_5F_0();audioEncoder_2E_blocks_2E_5_2E_mlp_5F_ln_2E_weight.bin(float32:512)*output:_2F_audioEncoder_2F_blocks_2E_5_2F_mlp_5F_ln_2F_Mul_5F_output_5F_0()
/audioEncoder/blocks.5/mlp_ln/Add_1:Add*input:_2F_audioEncoder_2F_blocks_2E_5_2F_mlp_5F_ln_2F_Mul_5F_output_5F_0();audioEncoder_2E_blocks_2E_5_2E_mlp_5F_ln_2E_bias.bin(float32:512)*output:_2F_audioEncoder_2F_blocks_2E_5_2F_mlp_5F_ln_2F_Add_5F_1_5F_output_5F_0()
/audioEncoder/blocks.5/mlp/mlp.0/MatMul:MatMul*input:_2F_audioEncoder_2F_blocks_2E_5_2F_mlp_5F_ln_2F_Add_5F_1_5F_output_5F_0();onnx_3A__3A_MatMul_5F_1358.bin(float32:512,2048)*output:_2F_audioEncoder_2F_blocks_2E_5_2F_mlp_2F_mlp_2E_0_2F_MatMul_5F_output_5F_0()
/audioEncoder/blocks.5/mlp/mlp.0/Add:Add*input:onnx_3A__3A_Add_5F_1357.bin(float32:2048);_2F_audioEncoder_2F_blocks_2E_5_2F_mlp_2F_mlp_2E_0_2F_MatMul_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_5_2F_mlp_2F_mlp_2E_0_2F_Add_5F_output_5F_0()
/audioEncoder/blocks.5/mlp/mlp.1/Div:Div*input:_2F_audioEncoder_2F_blocks_2E_5_2F_mlp_2F_mlp_2E_0_2F_Add_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_output_5F_0.bin(float32:)*output:_2F_audioEncoder_2F_blocks_2E_5_2F_mlp_2F_mlp_2E_1_2F_Div_5F_output_5F_0()
/audioEncoder/blocks.5/mlp/mlp.1/Erf:Erf*input:_2F_audioEncoder_2F_blocks_2E_5_2F_mlp_2F_mlp_2E_1_2F_Div_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_5_2F_mlp_2F_mlp_2E_1_2F_Erf_5F_output_5F_0()
/audioEncoder/blocks.5/mlp/mlp.1/Add:Add*input:_2F_audioEncoder_2F_blocks_2E_5_2F_mlp_2F_mlp_2E_1_2F_Erf_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_1_5F_output_5F_0.bin(float32:)*output:_2F_audioEncoder_2F_blocks_2E_5_2F_mlp_2F_mlp_2E_1_2F_Add_5F_output_5F_0()
/audioEncoder/blocks.5/mlp/mlp.1/Mul:Mul*input:_2F_audioEncoder_2F_blocks_2E_5_2F_mlp_2F_mlp_2E_0_2F_Add_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_5_2F_mlp_2F_mlp_2E_1_2F_Add_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_5_2F_mlp_2F_mlp_2E_1_2F_Mul_5F_output_5F_0()
/audioEncoder/blocks.5/mlp/mlp.1/Mul_1:Mul*input:_2F_audioEncoder_2F_blocks_2E_5_2F_mlp_2F_mlp_2E_1_2F_Mul_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_2_5F_output_5F_0.bin(float32:)*output:_2F_audioEncoder_2F_blocks_2E_5_2F_mlp_2F_mlp_2E_1_2F_Mul_5F_1_5F_output_5F_0()
/audioEncoder/blocks.5/mlp/mlp.2/MatMul:MatMul*input:_2F_audioEncoder_2F_blocks_2E_5_2F_mlp_2F_mlp_2E_1_2F_Mul_5F_1_5F_output_5F_0();onnx_3A__3A_MatMul_5F_1361.bin(float32:2048,512)*output:_2F_audioEncoder_2F_blocks_2E_5_2F_mlp_2F_mlp_2E_2_2F_MatMul_5F_output_5F_0()
/audioEncoder/blocks.5/mlp/mlp.2/Add:Add*input:onnx_3A__3A_Add_5F_1360.bin(float32:512);_2F_audioEncoder_2F_blocks_2E_5_2F_mlp_2F_mlp_2E_2_2F_MatMul_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_5_2F_mlp_2F_mlp_2E_2_2F_Add_5F_output_5F_0()
/audioEncoder/blocks.5/Add_1:Add*input:_2F_audioEncoder_2F_blocks_2E_5_2F_Add_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_5_2F_mlp_2F_mlp_2E_2_2F_Add_5F_output_5F_0()*output:_2F_audioEncoder_2F_blocks_2E_5_2F_Add_5F_1_5F_output_5F_0()
/audioEncoder/ln_post/ReduceMean:ReduceMean*input:_2F_audioEncoder_2F_blocks_2E_5_2F_Add_5F_1_5F_output_5F_0()*output:_2F_audioEncoder_2F_ln_5F_post_2F_ReduceMean_5F_output_5F_0()*axes:-1
/audioEncoder/ln_post/Sub:Sub*input:_2F_audioEncoder_2F_blocks_2E_5_2F_Add_5F_1_5F_output_5F_0();_2F_audioEncoder_2F_ln_5F_post_2F_ReduceMean_5F_output_5F_0()*output:_2F_audioEncoder_2F_ln_5F_post_2F_Sub_5F_output_5F_0()
/audioEncoder/ln_post/Pow:Pow*input:_2F_audioEncoder_2F_ln_5F_post_2F_Sub_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_0_2F_attn_5F_ln_2F_Constant_5F_output_5F_0.bin(float32:)*output:_2F_audioEncoder_2F_ln_5F_post_2F_Pow_5F_output_5F_0()
/audioEncoder/ln_post/ReduceMean_1:ReduceMean*input:_2F_audioEncoder_2F_ln_5F_post_2F_Pow_5F_output_5F_0()*output:_2F_audioEncoder_2F_ln_5F_post_2F_ReduceMean_5F_1_5F_output_5F_0()*axes:-1
/audioEncoder/ln_post/Add:Add*input:_2F_audioEncoder_2F_ln_5F_post_2F_ReduceMean_5F_1_5F_output_5F_0();_2F_audioEncoder_2F_blocks_2E_0_2F_attn_5F_ln_2F_Constant_5F_1_5F_output_5F_0.bin(float32:)*output:_2F_audioEncoder_2F_ln_5F_post_2F_Add_5F_output_5F_0()
/audioEncoder/ln_post/Sqrt:Sqrt*input:_2F_audioEncoder_2F_ln_5F_post_2F_Add_5F_output_5F_0()*output:_2F_audioEncoder_2F_ln_5F_post_2F_Sqrt_5F_output_5F_0()
/audioEncoder/ln_post/Div:Div*input:_2F_audioEncoder_2F_ln_5F_post_2F_Sub_5F_output_5F_0();_2F_audioEncoder_2F_ln_5F_post_2F_Sqrt_5F_output_5F_0()*output:_2F_audioEncoder_2F_ln_5F_post_2F_Div_5F_output_5F_0()
/audioEncoder/ln_post/Mul:Mul*input:_2F_audioEncoder_2F_ln_5F_post_2F_Div_5F_output_5F_0();audioEncoder_2E_ln_5F_post_2E_weight.bin(float32:512)*output:_2F_audioEncoder_2F_ln_5F_post_2F_Mul_5F_output_5F_0()
/audioEncoder/ln_post/Add_1:Add*input:_2F_audioEncoder_2F_ln_5F_post_2F_Mul_5F_output_5F_0();audioEncoder_2E_ln_5F_post_2E_bias.bin(float32:512)*output:_2F_audioEncoder_2F_ln_5F_post_2F_Add_5F_1_5F_output_5F_0()
/key/MatMul:MatMul*input:_2F_audioEncoder_2F_ln_5F_post_2F_Add_5F_1_5F_output_5F_0();onnx_3A__3A_MatMul_5F_1363.bin(float32:512,512)*output:_2F_key_2F_MatMul_5F_output_5F_0()
/value/MatMul:MatMul*input:_2F_audioEncoder_2F_ln_5F_post_2F_Add_5F_1_5F_output_5F_0();onnx_3A__3A_MatMul_5F_1366.bin(float32:512,512)*output:_2F_value_2F_MatMul_5F_output_5F_0()
/value/Add:Add*input:onnx_3A__3A_Add_5F_1365.bin(float32:512);_2F_value_2F_MatMul_5F_output_5F_0()*output:_2F_value_2F_Add_5F_output_5F_0()
/key_1/MatMul:MatMul*input:_2F_audioEncoder_2F_ln_5F_post_2F_Add_5F_1_5F_output_5F_0();onnx_3A__3A_MatMul_5F_1368.bin(float32:512,512)*output:_2F_key_5F_1_2F_MatMul_5F_output_5F_0()
/value_1/MatMul:MatMul*input:_2F_audioEncoder_2F_ln_5F_post_2F_Add_5F_1_5F_output_5F_0();onnx_3A__3A_MatMul_5F_1371.bin(float32:512,512)*output:_2F_value_5F_1_2F_MatMul_5F_output_5F_0()
/value_1/Add:Add*input:onnx_3A__3A_Add_5F_1370.bin(float32:512);_2F_value_5F_1_2F_MatMul_5F_output_5F_0()*output:_2F_value_5F_1_2F_Add_5F_output_5F_0()
/key_2/MatMul:MatMul*input:_2F_audioEncoder_2F_ln_5F_post_2F_Add_5F_1_5F_output_5F_0();onnx_3A__3A_MatMul_5F_1373.bin(float32:512,512)*output:_2F_key_5F_2_2F_MatMul_5F_output_5F_0()
/value_2/MatMul:MatMul*input:_2F_audioEncoder_2F_ln_5F_post_2F_Add_5F_1_5F_output_5F_0();onnx_3A__3A_MatMul_5F_1376.bin(float32:512,512)*output:_2F_value_5F_2_2F_MatMul_5F_output_5F_0()
/value_2/Add:Add*input:onnx_3A__3A_Add_5F_1375.bin(float32:512);_2F_value_5F_2_2F_MatMul_5F_output_5F_0()*output:_2F_value_5F_2_2F_Add_5F_output_5F_0()
/key_3/MatMul:MatMul*input:_2F_audioEncoder_2F_ln_5F_post_2F_Add_5F_1_5F_output_5F_0();onnx_3A__3A_MatMul_5F_1378.bin(float32:512,512)*output:_2F_key_5F_3_2F_MatMul_5F_output_5F_0()
/value_3/MatMul:MatMul*input:_2F_audioEncoder_2F_ln_5F_post_2F_Add_5F_1_5F_output_5F_0();onnx_3A__3A_MatMul_5F_1381.bin(float32:512,512)*output:_2F_value_5F_3_2F_MatMul_5F_output_5F_0()
/value_3/Add:Add*input:onnx_3A__3A_Add_5F_1380.bin(float32:512);_2F_value_5F_3_2F_MatMul_5F_output_5F_0()*output:_2F_value_5F_3_2F_Add_5F_output_5F_0()
/key_4/MatMul:MatMul*input:_2F_audioEncoder_2F_ln_5F_post_2F_Add_5F_1_5F_output_5F_0();onnx_3A__3A_MatMul_5F_1383.bin(float32:512,512)*output:_2F_key_5F_4_2F_MatMul_5F_output_5F_0()
/value_4/MatMul:MatMul*input:_2F_audioEncoder_2F_ln_5F_post_2F_Add_5F_1_5F_output_5F_0();onnx_3A__3A_MatMul_5F_1386.bin(float32:512,512)*output:_2F_value_5F_4_2F_MatMul_5F_output_5F_0()
/value_4/Add:Add*input:onnx_3A__3A_Add_5F_1385.bin(float32:512);_2F_value_5F_4_2F_MatMul_5F_output_5F_0()*output:_2F_value_5F_4_2F_Add_5F_output_5F_0()
/key_5/MatMul:MatMul*input:_2F_audioEncoder_2F_ln_5F_post_2F_Add_5F_1_5F_output_5F_0();onnx_3A__3A_MatMul_5F_1388.bin(float32:512,512)*output:_2F_key_5F_5_2F_MatMul_5F_output_5F_0()
/value_5/MatMul:MatMul*input:_2F_audioEncoder_2F_ln_5F_post_2F_Add_5F_1_5F_output_5F_0();onnx_3A__3A_MatMul_5F_1391.bin(float32:512,512)*output:_2F_value_5F_5_2F_MatMul_5F_output_5F_0()
/value_5/Add:Add*input:onnx_3A__3A_Add_5F_1390.bin(float32:512);_2F_value_5F_5_2F_MatMul_5F_output_5F_0()*output:_2F_value_5F_5_2F_Add_5F_output_5F_0()
/Unsqueeze:Unsqueeze*input:_2F_key_2F_MatMul_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1)*output:_2F_Unsqueeze_5F_output_5F_0()
/Unsqueeze_1:Unsqueeze*input:_2F_key_5F_1_2F_MatMul_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1)*output:_2F_Unsqueeze_5F_1_5F_output_5F_0()
/Unsqueeze_2:Unsqueeze*input:_2F_key_5F_2_2F_MatMul_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1)*output:_2F_Unsqueeze_5F_2_5F_output_5F_0()
/Unsqueeze_3:Unsqueeze*input:_2F_key_5F_3_2F_MatMul_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1)*output:_2F_Unsqueeze_5F_3_5F_output_5F_0()
/Unsqueeze_4:Unsqueeze*input:_2F_key_5F_4_2F_MatMul_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1)*output:_2F_Unsqueeze_5F_4_5F_output_5F_0()
/Unsqueeze_5:Unsqueeze*input:_2F_key_5F_5_2F_MatMul_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1)*output:_2F_Unsqueeze_5F_5_5F_output_5F_0()
/Concat:Concat*input:_2F_Unsqueeze_5F_output_5F_0();_2F_Unsqueeze_5F_1_5F_output_5F_0();_2F_Unsqueeze_5F_2_5F_output_5F_0();_2F_Unsqueeze_5F_3_5F_output_5F_0();_2F_Unsqueeze_5F_4_5F_output_5F_0();_2F_Unsqueeze_5F_5_5F_output_5F_0()*output:n_5F_layer_5F_cross_5F_k(6,0,0,512)*axis:0
/Unsqueeze_6:Unsqueeze*input:_2F_value_2F_Add_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1)*output:_2F_Unsqueeze_5F_6_5F_output_5F_0()
/Unsqueeze_7:Unsqueeze*input:_2F_value_5F_1_2F_Add_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1)*output:_2F_Unsqueeze_5F_7_5F_output_5F_0()
/Unsqueeze_8:Unsqueeze*input:_2F_value_5F_2_2F_Add_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1)*output:_2F_Unsqueeze_5F_8_5F_output_5F_0()
/Unsqueeze_9:Unsqueeze*input:_2F_value_5F_3_2F_Add_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1)*output:_2F_Unsqueeze_5F_9_5F_output_5F_0()
/Unsqueeze_10:Unsqueeze*input:_2F_value_5F_4_2F_Add_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1)*output:_2F_Unsqueeze_5F_10_5F_output_5F_0()
/Unsqueeze_11:Unsqueeze*input:_2F_value_5F_5_2F_Add_5F_output_5F_0();_2F_audioEncoder_2F_Constant_5F_7_5F_output_5F_0.bin(int64:1)*output:_2F_Unsqueeze_5F_11_5F_output_5F_0()
/Concat_1:Concat*input:_2F_Unsqueeze_5F_6_5F_output_5F_0();_2F_Unsqueeze_5F_7_5F_output_5F_0();_2F_Unsqueeze_5F_8_5F_output_5F_0();_2F_Unsqueeze_5F_9_5F_output_5F_0();_2F_Unsqueeze_5F_10_5F_output_5F_0();_2F_Unsqueeze_5F_11_5F_output_5F_0()*output:n_5F_layer_5F_cross_5F_v(6,0,0,512)*axis:0