Phase 2 upload: 529 clips

README.md

@@ -21,6 +21,108 @@ tags:
 - field-recording
 - whisper
 - wav2vec2
+dataset_info:
+  config_name: phase_2
+  features:
+  - name: audio
+    dtype:
+      audio:
+        sampling_rate: 16000
+  - name: duration_s
+    dtype: float64
+  - name: sample_rate
+    dtype: int64
+  - name: language_iso
+    dtype: string
+  - name: language_name
+    dtype: string
+  - name: language_family
+    dtype: string
+  - name: dialect
+    dtype: string
+  - name: country
+    dtype: string
+  - name: region
+    dtype: string
+  - name: gloss_en
+    dtype: string
+  - name: transcription_seke
+    dtype: string
+  - name: phonemes_ipa
+    dtype: string
+  - name: is_swadesh
+    dtype: bool
+  - name: swadesh_number
+    dtype: int64
+  - name: is_alternative_form
+    dtype: bool
+  - name: grammatical_categories
+    dtype: string
+  - name: source_identifier
+    dtype: string
+  - name: source_title
+    dtype: string
+  - name: source_filename
+    dtype: string
+  - name: linguistic_type
+    dtype: string
+  - name: recording_date
+    dtype: string
+  - name: recording_location
+    list: string
+  - name: speaker_name
+    dtype: string
+  - name: researcher_name
+    dtype: string
+  - name: project
+    dtype: string
+  - name: source_collection
+    dtype: string
+  - name: source_url
+    dtype: string
+  - name: license
+    dtype: string
+  - name: quality_tier
+    dtype: string
+  - name: snr_db
+    dtype: float64
+  - name: loudness_lufs
+    dtype: float64
+  - name: whisper_detected_lang
+    dtype: string
+  - name: whisper_avg_logprob
+    dtype: float64
+  - name: whisper_no_speech_prob
+    dtype: float64
+  splits:
+  - name: train
+    num_bytes: 25568035.0
+    num_examples: 423
+  - name: validation
+    num_bytes: 5528059.0
+    num_examples: 53
+  - name: test
+    num_bytes: 3622753.0
+    num_examples: 53
+  download_size: 32162046
+  dataset_size: 34718847.0
+configs:
+- config_name: default
+  data_files:
+  - split: test
+    path: data/test-*
+  - split: train
+    path: data/train-*
+  - split: validation
+    path: data/validation-*
+- config_name: phase_2
+  data_files:
+  - split: train
+    path: phase_2/train-*
+  - split: validation
+    path: phase_2/validation-*
+  - split: test
+    path: phase_2/test-*
 ---
 
 # 🏔️ Seke (SKJ) Language Dataset

phase_2/test-00000-of-00001.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:138d01b1ee7a49d20aae0ad2d985f1c2ad15eeca1460f92da50811aa30ab1363
+size 3603154

phase_2/train-00000-of-00001.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:83ec7e8633ae929727a9473e07b22eafcc713f60b30d42052ad4f6db95b8d043
+size 23222255

phase_2/validation-00000-of-00001.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e34293a77a86daeeee78d1f7e4157b2be180501700ae41f499037c3045924743
+size 5336637