Duplicate from ly-corporation/PASQA

Co-authored-by: LY Corporation <ly-ospo-ops@users.noreply.huggingface.co>

.gitattributes

@@ -0,0 +1,35 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,28 @@
+---
+license: apache-2.0
+language:
+- ja
+tags:
+- mos
+- speech-quality-assessment
+- pitch-accent
+- wav2vec2
+- audio
+---
+
+# PASQA — Pitch-Accent-Focused Speech Quality Assessment Model
+
+Pretrained checkpoint for **[PASQA](https://github.com/lycorp-jp/PASQA)**, an SSL-based MOS prediction model
+that estimates the perceptual quality of synthesized Japanese speech with a focus on
+**pitch-accent correctness**.
+
+The model is developed by LY Corporation and released alongside our INTERSPEECH 2026 paper.
+
+## Usage
+
+Inference is done with the code in the [PASQA](https://github.com/lycorp-jp/PASQA) repository. See the repository for setup and usage instructions.
+
+## License
+
+These model parameters are released under the
+[Apache License 2.0](https://www.apache.org/licenses/LICENSE-2.0).

checkpoint-100000steps.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:03c9e8880a28f65fd9b8611f3fe3e179020b067d892cd6f6a4c311572b8a8bc7
+size 753466693

config.yml

@@ -0,0 +1,32 @@
+model_input: waveform
+model_params:
+  attn_alpha_init: 0.1
+  attn_dim: 256
+  attn_dropout: 0.1
+  attn_heads: 4
+  error_head_dnn_dim: 64
+  mean_net_dnn_dim: 64
+  mean_net_output_type: scalar
+  mean_net_range_clipping: true
+  mora_dropout: 0.1
+  mora_emb_dim: 256
+  mora_ffn_dim: 512
+  mora_max_len: 128
+  mora_pos_encoding: rope
+  mora_transformer_heads: 4
+  mora_transformer_layers: 1
+  mora_vocab_size: 141
+  s3prl_name: wav2vec2
+  speaker_grl_dropout: 0.1
+  speaker_grl_hidden_dim: 128
+  ssl_model_layer_idx: -1
+  ssl_model_output_dim: 768
+  ssl_module: s3prl
+  use_error_head: true
+  use_listener_modeling: false
+  use_mean_listener: false
+  use_mora: true
+  use_speaker_grl: true
+mora_vocab_path: data/vocab.txt
+num_speakers: 13
+sampling_rate: 16000