diff --git a/.gitattributes b/.gitattributes
index a6344aac8c09253b3b630fb776ae94478aa0275b..349c955624180aeeb00ebb5dae7d72dce47180c2 100644
--- a/.gitattributes
+++ b/.gitattributes
@@ -33,3 +33,50 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+latent_diffusion/my_configs/audioldm_original/infer_07-23-08:28_cfg_scale_3.5_ddim_200_n_cand_3/-0xzrMun0Rs.wav filter=lfs diff=lfs merge=lfs -text
+testset_data/audiocaps/8Bu8CkR1xZY.wav filter=lfs diff=lfs merge=lfs -text
+testset_data/audiocaps/AHrUfa2H_5s.wav filter=lfs diff=lfs merge=lfs -text
+testset_data/audiocaps/XXBVsNt2Qr8.wav filter=lfs diff=lfs merge=lfs -text
+testset_data/audiocaps/3IYd8cCmUkQ.wav filter=lfs diff=lfs merge=lfs -text
+testset_data/audiocaps/hM88FG1_D5Q.wav filter=lfs diff=lfs merge=lfs -text
+testset_data/audiocaps/bl-eQ8XD5CY.wav filter=lfs diff=lfs merge=lfs -text
+testset_data/audiocaps/B90BOtSOD2Q.wav filter=lfs diff=lfs merge=lfs -text
+testset_data/audiocaps/6uIOGE36tWo.wav filter=lfs diff=lfs merge=lfs -text
+testset_data/audiocaps/HpkPTa1fQDE.wav filter=lfs diff=lfs merge=lfs -text
+testset_data/audiocaps/s72505MIhz8.wav filter=lfs diff=lfs merge=lfs -text
+testset_data/audiocaps/2gvyOxKuQPY.wav filter=lfs diff=lfs merge=lfs -text
+testset_data/audiocaps/bZuXMxR2S4U.wav filter=lfs diff=lfs merge=lfs -text
+testset_data/audiocaps/Cy3HWnwMLyI.wav filter=lfs diff=lfs merge=lfs -text
+testset_data/audiocaps/_gWEpDgPAho.wav filter=lfs diff=lfs merge=lfs -text
+testset_data/audiocaps/_u2cNlW5DxQ.wav filter=lfs diff=lfs merge=lfs -text
+testset_data/audiocaps/KpwdlYIdtfs.wav filter=lfs diff=lfs merge=lfs -text
+testset_data/audiocaps/EOaQnfDjVyo.wav filter=lfs diff=lfs merge=lfs -text
+testset_data/audiocaps/dTVkk4GSmNA.wav filter=lfs diff=lfs merge=lfs -text
+testset_data/audiocaps/ByOqw8M2U-Q.wav filter=lfs diff=lfs merge=lfs -text
+testset_data/audiocaps/xmaVWayBvaY.wav filter=lfs diff=lfs merge=lfs -text
+testset_data/audiocaps/LAHWV6fZwUk.wav filter=lfs diff=lfs merge=lfs -text
+testset_data/audiocaps/DW3z-ByrfWY.wav filter=lfs diff=lfs merge=lfs -text
+testset_data/audiocaps/gW33LYEvoaw.wav filter=lfs diff=lfs merge=lfs -text
+testset_data/audiocaps/7_Sr2zv1sQc.wav filter=lfs diff=lfs merge=lfs -text
+testset_data/audiocaps/fJ6ZeWYfLjA.wav filter=lfs diff=lfs merge=lfs -text
+testset_data/audiocaps/TkclVqlyKx4.wav filter=lfs diff=lfs merge=lfs -text
+testset_data/audiocaps/Fa1c4qfBqzE.wav filter=lfs diff=lfs merge=lfs -text
+testset_data/audiocaps/dPCj4WhTZ3c.wav filter=lfs diff=lfs merge=lfs -text
+testset_data/audiocaps/D3f5VIJYR7M.wav filter=lfs diff=lfs merge=lfs -text
+testset_data/audiocaps/3Xoz87_SUdw.wav filter=lfs diff=lfs merge=lfs -text
+testset_data/audiocaps/oxZjkE984Uo.wav filter=lfs diff=lfs merge=lfs -text
+testset_data/audiocaps/0Wdh45yt7tY.wav filter=lfs diff=lfs merge=lfs -text
+testset_data/audiocaps/A3vsvL-Yx0Q.wav filter=lfs diff=lfs merge=lfs -text
+testset_data/audiocaps/2RMOegT2Jn8.wav filter=lfs diff=lfs merge=lfs -text
+testset_data/audiocaps/jqiD3VeM_hY.wav filter=lfs diff=lfs merge=lfs -text
+testset_data/audiocaps/NSyqj1DXZKg.wav filter=lfs diff=lfs merge=lfs -text
+testset_data/audiocaps/NzVg-cFQJbE.wav filter=lfs diff=lfs merge=lfs -text
+testset_data/audiocaps/YwUa3OS92ZQ.wav filter=lfs diff=lfs merge=lfs -text
+testset_data/audiocaps/Bl-lCgr5hGY.wav filter=lfs diff=lfs merge=lfs -text
+testset_data/audiocaps/yrme-KRBvzk.wav filter=lfs diff=lfs merge=lfs -text
+testset_data/audiocaps/mFcHGbnNtSQ.wav filter=lfs diff=lfs merge=lfs -text
+testset_data/audiocaps/y6iMm7Pltq0.wav filter=lfs diff=lfs merge=lfs -text
+testset_data/audiocaps/xBDcJKb-9vk.wav filter=lfs diff=lfs merge=lfs -text
+testset_data/audiocaps/4mtfOkzOvBI.wav filter=lfs diff=lfs merge=lfs -text
+testset_data/audiocaps/05OJDYeHLMc.wav filter=lfs diff=lfs merge=lfs -text
+testset_data/audiocaps/VkXLtUx-RmI.wav filter=lfs diff=lfs merge=lfs -text
diff --git a/latent_diffusion/my_configs/audioldm_original/audioldm_original.yaml b/latent_diffusion/my_configs/audioldm_original/audioldm_original.yaml
new file mode 100644
index 0000000000000000000000000000000000000000..7ee2ed2715759557244cdbfb3370a8094a689c95
--- /dev/null
+++ b/latent_diffusion/my_configs/audioldm_original/audioldm_original.yaml
@@ -0,0 +1,149 @@
+metadata_root: "./data/metadata/dataset_root.json"
+log_directory: "./log/latent_diffusion"
+project: "audioldm"
+precision: "high"
+
+variables:
+  sampling_rate: &sampling_rate 16000 
+  mel_bins: &mel_bins 64
+  latent_embed_dim: &latent_embed_dim 8
+  latent_t_size: &latent_t_size 256 # TODO might need to change
+  latent_f_size: &latent_f_size 16
+  in_channels: &unet_in_channels 8
+  optimize_ddpm_parameter: &optimize_ddpm_parameter true
+  optimize_gpt: &optimize_gpt true
+  warmup_steps: &warmup_steps 2000
+
+data: 
+  train: ["audiocaps"]
+  val: "audiocaps"
+  test: "audiocaps"
+  class_label_indices: "audioset_eval_subset"
+  dataloader_add_ons: [] 
+
+step:
+  validation_every_n_epochs: 5
+  save_checkpoint_every_n_steps: 40000
+  max_steps: 800000
+  save_top_k: 2
+
+preprocessing:
+  audio:
+    sampling_rate: *sampling_rate
+    max_wav_value: 32768.0
+    duration: 10.24
+  stft:
+    filter_length: 1024
+    hop_length: 160
+    win_length: 1024
+  mel:
+    n_mel_channels: *mel_bins
+    mel_fmin: 0
+    mel_fmax: 8000 
+  num_workers: 2
+
+augmentation:
+  mixup: 0.0
+
+addition_params:
+  gradient_accumulation_steps: 1
+
+model:
+  target: audioldm_train.modules.latent_diffusion.ddpm.LatentDiffusion
+  params: 
+    # Autoencoder
+    first_stage_config:
+      base_learning_rate: 8.0e-06
+      target: audioldm_train.modules.latent_encoder.autoencoder.AutoencoderKL
+      params: 
+        reload_from_ckpt: "ckpts/vae_mel_16k_64bins.ckpt"
+        sampling_rate: *sampling_rate
+        batchsize: 3
+        monitor: val/rec_loss
+        image_key: fbank
+        subband: 1
+        embed_dim: *latent_embed_dim
+        time_shuffle: 1
+        lossconfig:
+          target: audioldm_train.losses.LPIPSWithDiscriminator
+          params:
+            disc_start: 50001
+            kl_weight: 1000.0
+            disc_weight: 0.5
+            disc_in_channels: 1
+        ddconfig: 
+          double_z: true
+          mel_bins: *mel_bins # The frequency bins of mel spectrogram
+          z_channels: 8
+          resolution: 256
+          downsample_time: false
+          in_channels: 1
+          out_ch: 1
+          ch: 128 
+          ch_mult:
+          - 1
+          - 2
+          - 4
+          num_res_blocks: 2
+          attn_resolutions: []
+          dropout: 0.0
+    
+    # Other parameters
+    base_learning_rate: 1.0e-4
+    warmup_steps: *warmup_steps
+    optimize_ddpm_parameter: *optimize_ddpm_parameter
+    sampling_rate: *sampling_rate
+    batchsize: 1
+    linear_start: 0.0015
+    linear_end: 0.0195
+    num_timesteps_cond: 1
+    log_every_t: 200
+    timesteps: 1000
+    unconditional_prob_cfg: 0.1
+    parameterization: eps # [eps, x0, v]
+    first_stage_key: fbank
+    latent_t_size: *latent_t_size # TODO might need to change
+    latent_f_size: *latent_f_size
+    channels: *latent_embed_dim # TODO might need to change
+    monitor: val/loss_simple_ema
+    scale_by_std: true
+    unet_config:
+      target: audioldm_train.modules.diffusionmodules.openaimodel.UNetModel
+      params:
+        image_size: 64 
+        extra_film_condition_dim: 512 # If you use film as extra condition, set this parameter. For example if you have two conditioning vectors each have dimension 512, then this number would be 1024
+        # context_dim: 
+        # - 768
+        in_channels: *unet_in_channels # The input channel of the UNet model
+        out_channels: *latent_embed_dim # TODO might need to change
+        model_channels: 128 # TODO might need to change
+        attention_resolutions:
+        - 8
+        - 4
+        - 2
+        num_res_blocks: 2
+        channel_mult: 
+        - 1
+        - 2
+        - 3
+        - 5
+        num_head_channels: 32
+        use_spatial_transformer: true
+        transformer_depth: 1
+        extra_sa_layer: false
+    
+    cond_stage_config:
+      film_clap_cond1:
+        cond_stage_key: text
+        conditioning_key: film
+        target: audioldm_train.conditional_models.CLAPAudioEmbeddingClassifierFreev2
+        params:
+          pretrained_path: ckpts/clap_htsat_tiny.pt
+          sampling_rate: 16000
+          embed_mode: text # or text
+          amodel: HTSAT-tiny
+
+    evaluation_params:
+      unconditional_guidance_scale: 3.5
+      ddim_sampling_steps: 200
+      n_candidates_per_samples: 3
diff --git a/latent_diffusion/my_configs/audioldm_original/checkpoints/checkpoint-fad-133.00-global_step=119999.ckpt b/latent_diffusion/my_configs/audioldm_original/checkpoints/checkpoint-fad-133.00-global_step=119999.ckpt
new file mode 100644
index 0000000000000000000000000000000000000000..c30d42c77631fd12e5460c3d4e9546dd918ff310
--- /dev/null
+++ b/latent_diffusion/my_configs/audioldm_original/checkpoints/checkpoint-fad-133.00-global_step=119999.ckpt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:81874520599c1cdc48f87af9a3d41ce228de409b9a31eae73e2c218819ee7feb
+size 4910952259
diff --git a/latent_diffusion/my_configs/audioldm_original/checkpoints/checkpoint-fad-133.00-global_step=79999.ckpt b/latent_diffusion/my_configs/audioldm_original/checkpoints/checkpoint-fad-133.00-global_step=79999.ckpt
new file mode 100644
index 0000000000000000000000000000000000000000..e33e0d15fe112ad89933b64c5525d757f9492612
--- /dev/null
+++ b/latent_diffusion/my_configs/audioldm_original/checkpoints/checkpoint-fad-133.00-global_step=79999.ckpt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cf8938853f65a559040b00d8164b58aa541396d807b5a6e3ba8c8ab9ef5285a9
+size 4910952259
diff --git a/latent_diffusion/my_configs/audioldm_original/infer_07-23-08:28_cfg_scale_3.5_ddim_200_n_cand_3/-0xzrMun0Rs.wav b/latent_diffusion/my_configs/audioldm_original/infer_07-23-08:28_cfg_scale_3.5_ddim_200_n_cand_3/-0xzrMun0Rs.wav
new file mode 100644
index 0000000000000000000000000000000000000000..0a09c9dbc9d738fed825e645360baff636f60654
--- /dev/null
+++ b/latent_diffusion/my_configs/audioldm_original/infer_07-23-08:28_cfg_scale_3.5_ddim_200_n_cand_3/-0xzrMun0Rs.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:325f313d578e4057eab5474cae4744464936581e7b71446bfbb003f96fc343eb
+size 327788
diff --git a/testset_data/audiocaps/05OJDYeHLMc.wav b/testset_data/audiocaps/05OJDYeHLMc.wav
new file mode 100644
index 0000000000000000000000000000000000000000..0ce2ac83b5d58b109b57182f140caf080069b8ae
--- /dev/null
+++ b/testset_data/audiocaps/05OJDYeHLMc.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cdb94d0d33f4b21375b3bef8d2ffd4ea063e4ad1bf08751cceacdb3ae3e7c400
+size 320044
diff --git a/testset_data/audiocaps/0Wdh45yt7tY.wav b/testset_data/audiocaps/0Wdh45yt7tY.wav
new file mode 100644
index 0000000000000000000000000000000000000000..6b62b1aaddfa24096fdc96f29928281e23eb8a09
--- /dev/null
+++ b/testset_data/audiocaps/0Wdh45yt7tY.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1f01912fdf5fae71dde63f09749570faf29c16ff8b289e3e23f88a681890cc45
+size 320044
diff --git a/testset_data/audiocaps/2RMOegT2Jn8.wav b/testset_data/audiocaps/2RMOegT2Jn8.wav
new file mode 100644
index 0000000000000000000000000000000000000000..b1f5cbbe6aea276bbb63116f9abdfd5d13ef0a36
--- /dev/null
+++ b/testset_data/audiocaps/2RMOegT2Jn8.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:020e2365d24b30c3ef632e9195bbe9dd56b643e6691bd3a6b44572df3bee92b2
+size 320044
diff --git a/testset_data/audiocaps/2gvyOxKuQPY.wav b/testset_data/audiocaps/2gvyOxKuQPY.wav
new file mode 100644
index 0000000000000000000000000000000000000000..32c271a7f0de50d19e2c33d11b54deb554f15949
--- /dev/null
+++ b/testset_data/audiocaps/2gvyOxKuQPY.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ef302a002971776b76428d76769766ae847f291f1ace33099b9e225f5ada710b
+size 320044
diff --git a/testset_data/audiocaps/3IYd8cCmUkQ.wav b/testset_data/audiocaps/3IYd8cCmUkQ.wav
new file mode 100644
index 0000000000000000000000000000000000000000..81bc0b159de350989ca3911f7b6aeb2f805d59ea
--- /dev/null
+++ b/testset_data/audiocaps/3IYd8cCmUkQ.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:85fb31647ed1e05f2eaf7bdede7f9097ee0cabad07fec6168b6d7154e102a5a6
+size 320044
diff --git a/testset_data/audiocaps/3Xoz87_SUdw.wav b/testset_data/audiocaps/3Xoz87_SUdw.wav
new file mode 100644
index 0000000000000000000000000000000000000000..849eecbfcd924454ef922ce77645b18fd36eb60d
--- /dev/null
+++ b/testset_data/audiocaps/3Xoz87_SUdw.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ee133da9f25dcb3a20a0299f732a39208be2c6f2546a5b14b5973db9f858b2f6
+size 320044
diff --git a/testset_data/audiocaps/4mtfOkzOvBI.wav b/testset_data/audiocaps/4mtfOkzOvBI.wav
new file mode 100644
index 0000000000000000000000000000000000000000..2aa83bb2bcca76bea3d4d92c1d1b79d965b8c324
--- /dev/null
+++ b/testset_data/audiocaps/4mtfOkzOvBI.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b245df371a41e0949fc0ef1181efaf1438e610d76348aa174303f19eb8a53065
+size 320044
diff --git a/testset_data/audiocaps/6uIOGE36tWo.wav b/testset_data/audiocaps/6uIOGE36tWo.wav
new file mode 100644
index 0000000000000000000000000000000000000000..c3aafe3dc7e8ffb47d8d82064a415436faadb417
--- /dev/null
+++ b/testset_data/audiocaps/6uIOGE36tWo.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:be77f7f2f24d4fa6b1b32e9b6bf78d3d391ec1ad9a0a3cc65e2f6c57ca497845
+size 320044
diff --git a/testset_data/audiocaps/7_Sr2zv1sQc.wav b/testset_data/audiocaps/7_Sr2zv1sQc.wav
new file mode 100644
index 0000000000000000000000000000000000000000..d5bf13f3f8933fc3f76e88988d9c2ada57ecab27
--- /dev/null
+++ b/testset_data/audiocaps/7_Sr2zv1sQc.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c3d1b94abf9d28803a5744b60e4de72bbe462aa817261ba8c3ee7a503d06a0db
+size 320044
diff --git a/testset_data/audiocaps/8Bu8CkR1xZY.wav b/testset_data/audiocaps/8Bu8CkR1xZY.wav
new file mode 100644
index 0000000000000000000000000000000000000000..03dc5d908e882a38bedbf7dfcbef0fa10c973ee6
--- /dev/null
+++ b/testset_data/audiocaps/8Bu8CkR1xZY.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f6f10d6da1e05c7978afa1b028fc755fe1d23d3796263d3878eb3ba83aaa085d
+size 320044
diff --git a/testset_data/audiocaps/A3vsvL-Yx0Q.wav b/testset_data/audiocaps/A3vsvL-Yx0Q.wav
new file mode 100644
index 0000000000000000000000000000000000000000..2a3ced44bb20375dc3cc88335eb8ccce782d1269
--- /dev/null
+++ b/testset_data/audiocaps/A3vsvL-Yx0Q.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f00c60d9e5e2b9071e8f2f898adaf78f7d392acf7274bf9b6738fffead31bf27
+size 320044
diff --git a/testset_data/audiocaps/AHrUfa2H_5s.wav b/testset_data/audiocaps/AHrUfa2H_5s.wav
new file mode 100644
index 0000000000000000000000000000000000000000..8e3a012626553d2118d1329748c324bb7a36a010
--- /dev/null
+++ b/testset_data/audiocaps/AHrUfa2H_5s.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5da7c879adac1e1624eb3483073f493fd0c837eadd497a45561850c5dad71eb5
+size 320044
diff --git a/testset_data/audiocaps/B90BOtSOD2Q.wav b/testset_data/audiocaps/B90BOtSOD2Q.wav
new file mode 100644
index 0000000000000000000000000000000000000000..ae6df4ae81e3783daf434ba2063a3468e449bdfa
--- /dev/null
+++ b/testset_data/audiocaps/B90BOtSOD2Q.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:54ef7017a184ee6c893bae9b484361e2c8d4acfabadfccca83ff7398fb5a8a22
+size 320044
diff --git a/testset_data/audiocaps/Bl-lCgr5hGY.wav b/testset_data/audiocaps/Bl-lCgr5hGY.wav
new file mode 100644
index 0000000000000000000000000000000000000000..5e2c6e59173aede1b6e87d1d0e96d98f666e0b42
--- /dev/null
+++ b/testset_data/audiocaps/Bl-lCgr5hGY.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c4ad3a9721ab468d3abecd3bfb33a04806dd7e83bf3ce60ad8eb1c5b2190722e
+size 320044
diff --git a/testset_data/audiocaps/ByOqw8M2U-Q.wav b/testset_data/audiocaps/ByOqw8M2U-Q.wav
new file mode 100644
index 0000000000000000000000000000000000000000..7befca7f2e684e295cc6efed2db1a89b3bb3ce55
--- /dev/null
+++ b/testset_data/audiocaps/ByOqw8M2U-Q.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c9dbbacf98fac39663c9ed5fb0b58302884c0ecb1375cdbc4cb18e6b8fd8e98c
+size 320044
diff --git a/testset_data/audiocaps/Cy3HWnwMLyI.wav b/testset_data/audiocaps/Cy3HWnwMLyI.wav
new file mode 100644
index 0000000000000000000000000000000000000000..8dc810d2e4479e56c4ef34435a59387d9e5a00ff
--- /dev/null
+++ b/testset_data/audiocaps/Cy3HWnwMLyI.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ccde8b6db494aafb5eae8bd2a69a6f9b381073b9ff242a89a9169ea249692a6c
+size 320044
diff --git a/testset_data/audiocaps/D3f5VIJYR7M.wav b/testset_data/audiocaps/D3f5VIJYR7M.wav
new file mode 100644
index 0000000000000000000000000000000000000000..7b78c6f3a46bf755615eeee4bbeef28213ea099d
--- /dev/null
+++ b/testset_data/audiocaps/D3f5VIJYR7M.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f0bf1d4c2bd7315270f311e409eb8e32e37bf591235d6feda184ad9cba0b5601
+size 320044
diff --git a/testset_data/audiocaps/DW3z-ByrfWY.wav b/testset_data/audiocaps/DW3z-ByrfWY.wav
new file mode 100644
index 0000000000000000000000000000000000000000..e984538ff72a5369a65392078c3717c827f57fb0
--- /dev/null
+++ b/testset_data/audiocaps/DW3z-ByrfWY.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:149f1afc2bb775a7c99ccb09c8e6e603b2a3ed5dd8c6c93028a9720bb1b4e99f
+size 320044
diff --git a/testset_data/audiocaps/EOaQnfDjVyo.wav b/testset_data/audiocaps/EOaQnfDjVyo.wav
new file mode 100644
index 0000000000000000000000000000000000000000..8d80abe104fd337351d4a6c28fb3d4901dfc3b79
--- /dev/null
+++ b/testset_data/audiocaps/EOaQnfDjVyo.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:40b92a82880c5c2ec12da32228887e0a721bd460cd3576ba21ff67bb127d4a8b
+size 320044
diff --git a/testset_data/audiocaps/Fa1c4qfBqzE.wav b/testset_data/audiocaps/Fa1c4qfBqzE.wav
new file mode 100644
index 0000000000000000000000000000000000000000..87eab8bf68e2dd6a0bcbb22af54780e589d69eed
--- /dev/null
+++ b/testset_data/audiocaps/Fa1c4qfBqzE.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:57a9d87e3781a8d74a97e6d9a28127aec0072c7ce34b9d4739f0e1a41cc28ccb
+size 320044
diff --git a/testset_data/audiocaps/HpkPTa1fQDE.wav b/testset_data/audiocaps/HpkPTa1fQDE.wav
new file mode 100644
index 0000000000000000000000000000000000000000..8e3b6b92a6ee154f4518548aaad37707515c4c02
--- /dev/null
+++ b/testset_data/audiocaps/HpkPTa1fQDE.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d3ae0d62cb8060e0a9101139c96df1867128f79182b35c7edb19b3e57295cdfe
+size 320044
diff --git a/testset_data/audiocaps/KpwdlYIdtfs.wav b/testset_data/audiocaps/KpwdlYIdtfs.wav
new file mode 100644
index 0000000000000000000000000000000000000000..9cb92bfdf27e45ad78d2c53603419b6c62f1cde9
--- /dev/null
+++ b/testset_data/audiocaps/KpwdlYIdtfs.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0280771fbe19535d4684992f9d3df3bdbfceebb670a7ebcb2e4834e4920b4843
+size 320044
diff --git a/testset_data/audiocaps/LAHWV6fZwUk.wav b/testset_data/audiocaps/LAHWV6fZwUk.wav
new file mode 100644
index 0000000000000000000000000000000000000000..907e0a1aa206fe9d6201f0406cc5a2e10a93ebd5
--- /dev/null
+++ b/testset_data/audiocaps/LAHWV6fZwUk.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:64ba52fb3c6fe92f0fd689cae48f5d189a00306ceda745004b2d79db7d70a949
+size 320044
diff --git a/testset_data/audiocaps/NSyqj1DXZKg.wav b/testset_data/audiocaps/NSyqj1DXZKg.wav
new file mode 100644
index 0000000000000000000000000000000000000000..fa486eb4458cc47f70650287b938e927ec2262f5
--- /dev/null
+++ b/testset_data/audiocaps/NSyqj1DXZKg.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d4ac4e9613525a966dd9da905529621422489ee17a4d789b1e05efefd8e3e01e
+size 320044
diff --git a/testset_data/audiocaps/NzVg-cFQJbE.wav b/testset_data/audiocaps/NzVg-cFQJbE.wav
new file mode 100644
index 0000000000000000000000000000000000000000..cb3b8eddc3c611b217487ff45e5581945a863f80
--- /dev/null
+++ b/testset_data/audiocaps/NzVg-cFQJbE.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8bfbd09f9483f395ea7b1a87dc7c3e18242ddf5f262a267ad54cc4241dfbd3db
+size 320044
diff --git a/testset_data/audiocaps/TkclVqlyKx4.wav b/testset_data/audiocaps/TkclVqlyKx4.wav
new file mode 100644
index 0000000000000000000000000000000000000000..10853facbc0ecd956ca053682a56d8c340803582
--- /dev/null
+++ b/testset_data/audiocaps/TkclVqlyKx4.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9da3ca97bf3fa79478daccf8eb924062eca8af88dc99845c0827d9cbe78fa7fc
+size 320044
diff --git a/testset_data/audiocaps/VkXLtUx-RmI.wav b/testset_data/audiocaps/VkXLtUx-RmI.wav
new file mode 100644
index 0000000000000000000000000000000000000000..336396180d9b8d7ed9936e953508112f2805d7df
--- /dev/null
+++ b/testset_data/audiocaps/VkXLtUx-RmI.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:640f780ea1c043641462342fef56e9db077b0aaed1a636e9dd61b71c0c83d53f
+size 320044
diff --git a/testset_data/audiocaps/XXBVsNt2Qr8.wav b/testset_data/audiocaps/XXBVsNt2Qr8.wav
new file mode 100644
index 0000000000000000000000000000000000000000..66debb4341afc25b72501a8adeb9b4b97b63e047
--- /dev/null
+++ b/testset_data/audiocaps/XXBVsNt2Qr8.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5100bfe43151f773135294c6fa17a77cc92a1f901cb84c1c5be75358e7618645
+size 320044
diff --git a/testset_data/audiocaps/YwUa3OS92ZQ.wav b/testset_data/audiocaps/YwUa3OS92ZQ.wav
new file mode 100644
index 0000000000000000000000000000000000000000..7d8c68a054750aa068a5638979dc9f7793490e5a
--- /dev/null
+++ b/testset_data/audiocaps/YwUa3OS92ZQ.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:031d0952cec730868e7ad3c290005ed0fe55470cdb3e029545768b4605fb88c7
+size 320044
diff --git a/testset_data/audiocaps/_gWEpDgPAho.wav b/testset_data/audiocaps/_gWEpDgPAho.wav
new file mode 100644
index 0000000000000000000000000000000000000000..e66bf3e013f86ceb2629eddea215070c515713b5
--- /dev/null
+++ b/testset_data/audiocaps/_gWEpDgPAho.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3bec87ac5b40c2d9ebf26118b12d4f90a4f48e3766608ba4ed27f05c63e31c7b
+size 320044
diff --git a/testset_data/audiocaps/_u2cNlW5DxQ.wav b/testset_data/audiocaps/_u2cNlW5DxQ.wav
new file mode 100644
index 0000000000000000000000000000000000000000..fab20d05c0500bd0d00538e0e8c3550836691b3c
--- /dev/null
+++ b/testset_data/audiocaps/_u2cNlW5DxQ.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b2bf139665d39580e62359d131938a1f260219a3cc0e1a279f0c107e588fb145
+size 320044
diff --git a/testset_data/audiocaps/bZuXMxR2S4U.wav b/testset_data/audiocaps/bZuXMxR2S4U.wav
new file mode 100644
index 0000000000000000000000000000000000000000..f2e54cfc6bdbc885aecf9503541cbdba11677098
--- /dev/null
+++ b/testset_data/audiocaps/bZuXMxR2S4U.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fee3ac0c043fd0e9245b202a92cab10d3b7d3ddb585e9f1e7640e34c93d500d3
+size 320044
diff --git a/testset_data/audiocaps/bl-eQ8XD5CY.wav b/testset_data/audiocaps/bl-eQ8XD5CY.wav
new file mode 100644
index 0000000000000000000000000000000000000000..174d1d24f36c1336ae0dfc92dc5056fbb960f4b3
--- /dev/null
+++ b/testset_data/audiocaps/bl-eQ8XD5CY.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f569b41c49db3364ab5284f6d20c92046afb386501df08fc865e548e58a5cb6e
+size 320044
diff --git a/testset_data/audiocaps/dPCj4WhTZ3c.wav b/testset_data/audiocaps/dPCj4WhTZ3c.wav
new file mode 100644
index 0000000000000000000000000000000000000000..6ef746ccfa4640f90d99f9b11972ee58a77b9521
--- /dev/null
+++ b/testset_data/audiocaps/dPCj4WhTZ3c.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f83cbd8264ded5bb2d573d73d318e6fb1a36988b288e96a4ef82312ea8b5b3e9
+size 320044
diff --git a/testset_data/audiocaps/dTVkk4GSmNA.wav b/testset_data/audiocaps/dTVkk4GSmNA.wav
new file mode 100644
index 0000000000000000000000000000000000000000..34437bfa4b6105fd3c636c56169852a83598c878
--- /dev/null
+++ b/testset_data/audiocaps/dTVkk4GSmNA.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c79ad7f36214e686c35ae3d234e865862190ed6e76c8d0eec996ff5efe64939f
+size 320044
diff --git a/testset_data/audiocaps/fJ6ZeWYfLjA.wav b/testset_data/audiocaps/fJ6ZeWYfLjA.wav
new file mode 100644
index 0000000000000000000000000000000000000000..f08d11053b106febbc4a81e2e39c4aec696d375b
--- /dev/null
+++ b/testset_data/audiocaps/fJ6ZeWYfLjA.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3954307043dd101f3a270cccdfd59a960a4a7c3e5df478881ca0375049a1d87d
+size 320044
diff --git a/testset_data/audiocaps/gW33LYEvoaw.wav b/testset_data/audiocaps/gW33LYEvoaw.wav
new file mode 100644
index 0000000000000000000000000000000000000000..251e5876cb6e5a89879c3e11f98dc4499cbb7754
--- /dev/null
+++ b/testset_data/audiocaps/gW33LYEvoaw.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c12539607e235125526aef365e551a45b8b5212f27f9988b6f9617a20c891039
+size 320044
diff --git a/testset_data/audiocaps/hM88FG1_D5Q.wav b/testset_data/audiocaps/hM88FG1_D5Q.wav
new file mode 100644
index 0000000000000000000000000000000000000000..70713fa99d96732b49a1a68cacef6b9c6c4c178b
--- /dev/null
+++ b/testset_data/audiocaps/hM88FG1_D5Q.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f0babba6248fc32bb99d85fd90e0f3beb3b84fa8d86f45e7f5cf1e28b5a35f3c
+size 320044
diff --git a/testset_data/audiocaps/jqiD3VeM_hY.wav b/testset_data/audiocaps/jqiD3VeM_hY.wav
new file mode 100644
index 0000000000000000000000000000000000000000..a71c7f358dafd6384a42f81eedf8dc742d23b3b3
--- /dev/null
+++ b/testset_data/audiocaps/jqiD3VeM_hY.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8473f523e5cf35621318aa382c47ad5863df18d9409e81c859440e6a928793e2
+size 320044
diff --git a/testset_data/audiocaps/mFcHGbnNtSQ.wav b/testset_data/audiocaps/mFcHGbnNtSQ.wav
new file mode 100644
index 0000000000000000000000000000000000000000..899c7bfcc2c17e6d461ce6d3a182d00863d772f1
--- /dev/null
+++ b/testset_data/audiocaps/mFcHGbnNtSQ.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:644b53e6bdf01ef750911911acca3b492f04a966418c93072695864a06aefdb1
+size 320044
diff --git a/testset_data/audiocaps/oxZjkE984Uo.wav b/testset_data/audiocaps/oxZjkE984Uo.wav
new file mode 100644
index 0000000000000000000000000000000000000000..42ae1132b256b0c4c9bc46f68997aa2c3f03e405
--- /dev/null
+++ b/testset_data/audiocaps/oxZjkE984Uo.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d1c619b9b340a904e7f3026be08a0741a74c075282aaa7cedf29f6a448a890ae
+size 320044
diff --git a/testset_data/audiocaps/s72505MIhz8.wav b/testset_data/audiocaps/s72505MIhz8.wav
new file mode 100644
index 0000000000000000000000000000000000000000..d4c95d825c9764d872d8460f4c7e3b92bfbcccc8
--- /dev/null
+++ b/testset_data/audiocaps/s72505MIhz8.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5eea74f52099773a9db2444cae186a025d5950c729b8d4cc604ba18fd0f882de
+size 320044
diff --git a/testset_data/audiocaps/xBDcJKb-9vk.wav b/testset_data/audiocaps/xBDcJKb-9vk.wav
new file mode 100644
index 0000000000000000000000000000000000000000..27717de1cbadc8b1622ebeb33386ce045b4f03dd
--- /dev/null
+++ b/testset_data/audiocaps/xBDcJKb-9vk.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d0d5718a1c5f233d0639a098e008f373724710d856723c3a2b3e5ceeb38628ca
+size 320044
diff --git a/testset_data/audiocaps/xmaVWayBvaY.wav b/testset_data/audiocaps/xmaVWayBvaY.wav
new file mode 100644
index 0000000000000000000000000000000000000000..261f2eaa3163de723a34b7451f0eb927e00ce700
--- /dev/null
+++ b/testset_data/audiocaps/xmaVWayBvaY.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:873f54931176cc29d03ca870612ebb6ba1cb55549e3c07cfc8cd30a45d9fe1d8
+size 320044
diff --git a/testset_data/audiocaps/y6iMm7Pltq0.wav b/testset_data/audiocaps/y6iMm7Pltq0.wav
new file mode 100644
index 0000000000000000000000000000000000000000..5615ea1d626fc7e2c0caef23accb231074d426b2
--- /dev/null
+++ b/testset_data/audiocaps/y6iMm7Pltq0.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a3092ec17275e2a16280d09d6cc5e27c8c0571f366abdb491ff7844941cc208
+size 320044
diff --git a/testset_data/audiocaps/yrme-KRBvzk.wav b/testset_data/audiocaps/yrme-KRBvzk.wav
new file mode 100644
index 0000000000000000000000000000000000000000..ad0f2a05187b8df50161c21eade8ab58b8a304a1
--- /dev/null
+++ b/testset_data/audiocaps/yrme-KRBvzk.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:994c75b2ad6cdd0b21f5d0e08177c0e61b7bd5837acf88491bb1a9ec5ab6a8d3
+size 320044