Add evaluation results with quality, accuracy, and speaker similarity metrics

README.md

@@ -0,0 +1,35 @@
+---
+dataset_info:
+  features:
+  - name: text
+    dtype: string
+  - name: model
+    dtype: string
+  - name: MOS
+    dtype: float64
+  - name: NOI
+    dtype: float64
+  - name: DIS
+    dtype: float64
+  - name: COL
+    dtype: float64
+  - name: LOUD
+    dtype: float64
+  - name: UTMOS
+    dtype: float64
+  - name: WER
+    dtype: float64
+  - name: CER
+    dtype: float64
+  splits:
+  - name: train
+    num_bytes: 188430
+    num_examples: 1440
+  download_size: 87645
+  dataset_size: 188430
+configs:
+- config_name: default
+  data_files:
+  - split: train
+    path: data/train-*
+---

data/train-00000-of-00001.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9fe9efc351e01e640afee700b3c4940ecc16ef87c4f4ed08b6606ac81f33fce1
+size 87645