Update advanced instructions

README.md

@@ -1,5 +1,5 @@
 ---
-# Data-files config (no script, Parquet-only)
+# Data-files config (Parquet-only)
 configs:
 - config_name: basic
   data_files:
@@ -13,7 +13,14 @@ configs:
       - "basic/test.parquet"
   default: true
 
-pretty_name: "Gametime (Basic)"
+- config_name: advanced
+  data_files:
+  - split: test
+    path:
+      - "advanced/test-*.parquet"
+      - "advanced/test.parquet"
+
+pretty_name: "Gametime"
 tags:
   - audio
   - speech
@@ -23,6 +30,7 @@ tags:
 task_categories:
   - automatic-speech-recognition
   - text-to-speech
+  - language-modeling
 language:
   - en
 license: cc-by-4.0
@@ -30,27 +38,30 @@ size_categories:
   - n<100K
 ---
 
-# Gametime — Basic (Parquet)
+# Gametime Benchmark
 
-The **Basic** split of Gametime is a lightweight, streaming-friendly dataset for quick TTS/ASR prototyping.  
-Audio is stored directly as raw bytes in Parquet (no `Audio` feature), so you can stream and decode without extra dependencies.
+The **Gametime** dataset provides lightweight, streaming-friendly splits for TTS/ASR/SpokenLM prototyping.
 
 ---
 
 ## 📦 Download Options
 
 ### 1️⃣ Recommended — Stream from Hugging Face
+
 ```python
 from datasets import load_dataset
 import io, soundfile as sf
 
-ds = load_dataset("gametime-benchmark/gametime", "basic", split="train", streaming=True)
+# Load Basic train split
+ds_basic = load_dataset("gametime-benchmark/gametime", "basic", split="train", streaming=True)
+ex = next(iter(ds_basic))
+print(ex["id"], ex["text"])
 
-ex = next(iter(ds))
-buf = io.BytesIO(ex["audio_bytes"])
-wav, sr = sf.read(buf, dtype="float32")
-print(ex["id"], sr, len(wav), ex["text"])
-````
+# Load Advanced test split
+ds_adv = load_dataset("gametime-benchmark/gametime", "advanced", split="test", streaming=True)
+ex_adv = next(iter(ds_adv))
+print(ex_adv["id"], ex_adv["text"])
+```
 
 * Works with **`streaming=True`** — no full download needed
 * Requires only `soundfile` (libsndfile)
@@ -63,45 +74,37 @@ If you prefer the original folder layout:
 
 ```
 gametime/basic/
-├── 1-a-Sequence-Number-dataset.json
 ├── ...*-dataset.json
 ├── audios/
-│   ├── 1-a-Sequence-Number/
-│   │   ├── train/*.wav
-│   │   ├── test/*.wav
-...
-```
+│   ├── .../*.wav
 
-You can download and unzip:
+gametime/advanced/
+├── ...*-dataset.json
+├── audios/
+│   ├── test/*.wav
+```
 
 ```python
-# pip install -U huggingface_hub
-import os
 from huggingface_hub import hf_hub_download
-try:
-    import dotenv
-    dotenv.load_dotenv()  # Load .env if available
-except:
-    print("dotenv not installed, skipping .env loading")
+import os
 
 token = os.getenv("HF_TOKEN") or "<YOUR_HF_TOKEN>"
-
 path = hf_hub_download(
     repo_id="gametime-benchmark/gametime",
     repo_type="dataset",
-    filename="download/gametime_basic_v1.zip", 
-    revision="main", 
+    filename="download/gametime_advanced_v1.zip",
+    revision="main",
     token=token,
-    local_dir=".",     # download to current directory (change as needed)
+    local_dir=".",
 )
 print("saved to:", path)
 ```
 
 ```bash
-unzip download/gametime_basic_v1.zip
+unzip download/gametime_advanced_v1.zip
 ```
 
-This preserves the `gametime/basic` directory tree exactly as in source.
+This preserves the `gametime/advanced` directory tree exactly as in source.
 
 ---
 
@@ -112,7 +115,7 @@ Each Parquet row has:
 | Column          | Type  | Description                                                    |
 | --------------- | ----- | -------------------------------------------------------------- |
 | `id`            | str   | e.g. `1-a-Sequence-Number/train/1-a-Sequence-Number-01-01.wav` |
-| `category`      | str   | `"basic"`                                                      |
+| `category`      | str   | `"basic"` or `"advanced"`                                      |
 | `dataset`       | str   | group name (e.g. `1-a-Sequence-Number`)                        |
 | `split`         | str   | `train` or `test`                                              |
 | `template_idx`  | str   | template index if available                                    |
@@ -137,7 +140,8 @@ from transformers import pipeline
 asr = pipeline("automatic-speech-recognition", model="openai/whisper-small")
 wer = evaluate.load("wer")
 
-ds = load_dataset("gametime-benchmark/gametime", "basic", split="test", streaming=True)
+# Evaluate Advanced split (test only)
+ds = load_dataset("gametime-benchmark/gametime", "advanced", split="test", streaming=True)
 
 refs, hyps = [], []
 for i, ex in enumerate(ds):
@@ -145,7 +149,7 @@ for i, ex in enumerate(ds):
     hyp = asr({"array": wav, "sampling_rate": sr})["text"]
     refs.append(ex["text"]); hyps.append(hyp)
     print(f"Sample {i+1}: {ex['text']} -> {hyp}")
-    if i >= 199:  # quick 200-sample check
+    if i >= 199:
         break
 
 print("WER:", wer.compute(references=refs, predictions=hyps))
@@ -155,7 +159,8 @@ print("WER:", wer.compute(references=refs, predictions=hyps))
 
 ## 📝 Notes
 
-* **Advanced** split (with `<sil="Xms"/>` timing tokens) is planned separately.
+* **Basic**: clean split with plain text transcripts.
+* **Advanced**: adds inline timing tokens such as `<sil="Xms"/>`, enabling alignment-aware training and evaluation.
 * License: **CC BY 4.0** for text, metadata, and audio.
 * ZIP download is optional; Parquet streaming is the default recommended method.
 
@@ -164,10 +169,7 @@ print("WER:", wer.compute(references=refs, predictions=hyps))
 ## 📚 Citation
 
 ```
-@dataset{gametime_basic_2025,
-  title   = {Gametime: Basic Speech Benchmark},
-  year    = {2025},
-  url     = {https://huggingface.co/datasets/gametime-benchmark/gametime},
-  note    = {Parquet-only split "basic" for TTS/ASR sanity checks}
-}
+(TO BE UPDATED)
 ```
+
+---

advanced/test.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0b190b4ead73d9c1924bce91b409f7c6d2938d82a5320574dd85151691452e2d
+size 481376638