Upload README.md with huggingface_hub

README.md

@@ -41,8 +41,11 @@ This dataset is derived from [LibriSpeech](https://huggingface.co/datasets/opens
 
 | Feature | Type | Description |
 |---------|------|-------------|
-| `input_values` | `Sequence[float]` | Normalized audio waveform (16kHz) |
+| `audio` | `Audio` | Original audio (FLAC, 16kHz) - playable in HF viewer |
+| `input_values` | `Sequence[float]` | Normalized audio waveform (float32) - ready for training |
 | `labels` | `Sequence[int]` | ARPAbet phoneme token IDs |
+| `text` | `string` | Original text transcription |
+| `duration` | `float` | Audio duration in seconds |
 
 ### ARPAbet Vocabulary (72 tokens)
 
@@ -66,9 +69,18 @@ from datasets import load_dataset
 from huggingface_hub import hf_hub_download
 import json
 
-# Load the dataset
+# Load the full dataset
 dataset = load_dataset("davidggphy/librispeech-arpabet-processed")
 
+# Or load a subset (faster for testing)
+dataset = load_dataset("davidggphy/librispeech-arpabet-processed", split="train[:1000]")  # First 1000
+dataset = load_dataset("davidggphy/librispeech-arpabet-processed", split="train[:10%]")   # First 10%
+
+# Streaming mode (no full download)
+stream = load_dataset("davidggphy/librispeech-arpabet-processed", split="train", streaming=True)
+for sample in stream.take(100):
+    print(sample["text"])
+
 # Load vocabulary mapping
 vocab_path = hf_hub_download(
     repo_id="davidggphy/librispeech-arpabet-processed",
@@ -82,11 +94,37 @@ token_to_id = vocab_data["token_to_id"]
 id_to_token = {int(k): v for k, v in vocab_data["id_to_token"].items()}
 
 # Access samples
-sample = dataset["train"][0]
-print(f"Audio length: {len(sample['input_values'])} samples")
+sample = dataset[0]
+print(f"Text: {sample['text']}")
+print(f"Duration: {sample['duration']:.2f}s")
 print(f"Labels: {[id_to_token[i] for i in sample['labels']]}")
 ```
 
+### Listening to Audio
+
+The `input_values` column contains normalized audio waveforms at 16kHz. To play or save the audio:
+
+```python
+import numpy as np
+import soundfile as sf
+from IPython.display import Audio
+
+sample = dataset["train"][0]
+
+# Convert to numpy array
+audio = np.array(sample["input_values"], dtype=np.float32)
+
+# Play in Jupyter/Colab
+Audio(audio, rate=16000)
+
+# Or save to file
+sf.write("sample.wav", audio, 16000)
+
+# Check duration
+print(f"Duration: {sample['duration']:.2f}s")
+print(f"Text: {sample['text']}")
+```
+
 ### Training with Wav2Vec2
 
 ```python