add playable audio+text parquet (52512 paired, 6 shards)

README.md

@@ -18,38 +18,36 @@ configs:
   - config_name: default
     data_files:
       - split: train
-        path: metadata.jsonl
+        path: data/train-*.parquet
 ---
 
 # kupe-tts
 
-Hindi / Hinglish TTS pairs: **audio** + **text** (Soniox realtime TTS).
+Hindi / Hinglish TTS pairs: **playable audio** + **text** (Soniox realtime TTS).
 
 ## Dataset Viewer
 
-`metadata.jsonl` links each wav to its transcript so the Hub shows both columns:
+Default split is parquet with a real **Audio** feature (player in the Hub UI):
 
-| column | description |
-|--------|-------------|
-| `file_name` | path under `audio/` (flat or sharded) |
-| `text` | utterance transcript from the corpus |
-| `index` | corpus index |
+| column | type | description |
+|--------|------|-------------|
+| `audio` | Audio | playable wav already stored under `audio/` |
+| `text` | string | utterance transcript |
+| `index` | int64 | corpus index |
 
-Paired rows in this card build: **52,512** (audio present **and** non-empty text).
+Paired rows: **52,512** (only rows with **both** Hub audio and non-empty text).
 
 ```python
 from datasets import load_dataset
 ds = load_dataset("anuj-inavlabs/kupe-tts", split="train")
-# or AudioFolder:
-# ds = load_dataset("audiofolder", data_files={"train": "metadata.jsonl"}, ...)
-print(ds[0]["text"], ds[0]["file_name"])
+print(ds[0]["text"])
+# ds[0]["audio"] → array / sampling_rate / path
 ```
 
 ## Layout
 
+- `data/train-*.parquet` — Viewer source (`audio` + `text` + `index`)
 - `audio/` — wav files (`audio/NNNNNN.wav` or `audio/SSS/NNNNNN.wav`)
 - `timestamps/` — Soniox char-level timestamps JSON
-- `metadata.jsonl` — **required for text column** in the Dataset Viewer
+- `metadata.jsonl` — AudioFolder-style side index (`file_name` + `text`)
 - `tracker/` — generation / cost trackers
-
-Only rows with **both** audio and text are listed in `metadata.jsonl`.

data/train-00000-of-00006.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:18f3beaebb591c3f6ffa12d4b23593ec58eb9bdf9a7b10f1cae2f2ece2e7bd85
+size 353920

data/train-00001-of-00006.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:77d1dbdd61be56a0b413459dd521344efa23fa9808741db90f2855fe987b8ee1
+size 332931

data/train-00002-of-00006.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f15a8e60446b235fca1fe6e194ab895ddd775175a72fa1790a3c01cbb30e9bca
+size 435050

data/train-00003-of-00006.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bc6568260c67e0818954ea09eaeeea6c1661145f638d82383b400029506d752a
+size 433903

data/train-00004-of-00006.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:82aaa17272b5203ee55ab3eb081ff956bb62c29bebee63a369e9ca7432b0ff36
+size 434208

data/train-00005-of-00006.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:271b53b38ffdc4049636ca8ffbb3b591ca3e04f54f98fbc90df65802d4c286de
+size 114029