Fix metadata format for Hugging Face Dataset Viewer compatibility

- Updated metadata.jsonl to use file_name column instead of audio field
- Restructured data with audio files in data/audio/ directory
- Updated all file paths to match new directory structure
- Modified README examples to reflect new metadata format
- Created test scripts to verify HF audio dataset compatibility

This follows HF audio dataset standards to ensure word and IPA columns
are visible in the Dataset Viewer alongside audio files.

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>

README.md

@@ -52,9 +52,9 @@ Each instance contains:
 
 ### Data Fields
 
+- `file_name` (Audio): Audio file path that will be automatically loaded by Hugging Face Datasets
 - `word` (string): Vietnamese word or phrase
 - `ipa` (string): IPA transcription with tone marks
-- `audio` (Audio): Audio file path that will be automatically loaded by Hugging Face Datasets
 
 ### Data Splits
 
@@ -105,8 +105,9 @@ dataset = load_dataset("undertheseanlp/uts2025_vietipa")
 for example in dataset['train']:
     word = example['word']
     ipa = example['ipa']
-    audio = example['audio']  # Audio object with array, path, and sampling_rate
+    audio = example['file_name']  # Audio object with array, path, and sampling_rate
     print(f"{word}: {ipa}")
+    print(f"Audio path: {audio['path']}")
     print(f"Audio sampling rate: {audio['sampling_rate']} Hz")
 ```
 
@@ -114,7 +115,6 @@ for example in dataset['train']:
 
 ```python
 from datasets import load_dataset
-import soundfile as sf
 
 # Load the dataset
 dataset = load_dataset("undertheseanlp/uts2025_vietipa")
@@ -123,7 +123,7 @@ dataset = load_dataset("undertheseanlp/uts2025_vietipa")
 for example in dataset['train'].select(range(3)):  # First 3 examples
     word = example['word']
     ipa = example['ipa']
-    audio = example['audio']
+    audio = example['file_name']
 
     print(f"Word: {word}")
     print(f"IPA: {ipa}")
@@ -158,22 +158,22 @@ uv run python example_usage.py
 ### Sample JSONL Records
 
 ```json
-{"word": "xin chào", "ipa": "sin˧˧ caːw˨˩", "audio": "audio/xin_chao.mp3"}
-{"word": "cảm ơn", "ipa": "kaːm˧˩ ʔəːn˧˧", "audio": "audio/cam_on.mp3"}
-{"word": "Việt Nam", "ipa": "viət˧˩ˀ naːm˧˧", "audio": "audio/viet_nam.mp3"}
-{"word": "nước", "ipa": "nɨək˧˥", "audio": "audio/nuoc.mp3"}
-{"word": "yêu", "ipa": "iəw˧˧", "audio": "audio/yeu.mp3"}
+{"file_name": "data/audio/xin_chao.mp3", "word": "xin chào", "ipa": "sin˧˧ caːw˨˩"}
+{"file_name": "data/audio/cam_on.mp3", "word": "cảm ơn", "ipa": "kaːm˧˩ ʔəːn˧˧"}
+{"file_name": "data/audio/viet_nam.mp3", "word": "Việt Nam", "ipa": "viət˧˩ˀ naːm˧˧"}
+{"file_name": "data/audio/nuoc.mp3", "word": "nước", "ipa": "nɨək˧˥"}
+{"file_name": "data/audio/yeu.mp3", "word": "yêu", "ipa": "iəw˧˧"}
 ```
 
 ### Data Table View
 
-| Word | IPA | Audio File |
-|------|-----|------------|
-| xin chào | sin˧˧ caːw˨˩ | audio/xin_chao.mp3 |
-| cảm ơn | kaːm˧˩ ʔəːn˧˧ | audio/cam_on.mp3 |
-| Việt Nam | viət˧˩ˀ naːm˧˧ | audio/viet_nam.mp3 |
-| nước | nɨək˧˥ | audio/nuoc.mp3 |
-| yêu | iəw˧˧ | audio/yeu.mp3 |
+| Audio File | Word | IPA |
+|------------|------|-----|
+| data/audio/xin_chao.mp3 | xin chào | sin˧˧ caːw˨˩ |
+| data/audio/cam_on.mp3 | cảm ơn | kaːm˧˩ ʔəːn˧˧ |
+| data/audio/viet_nam.mp3 | Việt Nam | viət˧˩ˀ naːm˧˧ |
+| data/audio/nuoc.mp3 | nước | nɨək˧˥ |
+| data/audio/yeu.mp3 | yêu | iəw˧˧ |
 
 ## Considerations for Using the Data
 

convert_to_hf_audio.py

@@ -1,7 +1,6 @@
 #!/usr/bin/env python3
 
 import json
-from pathlib import Path
 
 
 def convert_to_hf_audio_format(jsonl_file="data/metadata.jsonl", output_file="data/metadata_hf.jsonl"):

convert_to_jsonl.py

@@ -2,7 +2,6 @@
 
 import csv
 import json
-from pathlib import Path
 
 
 def convert_csv_to_jsonl(csv_file="vietnamese_ipa_dataset.csv", jsonl_file="vietnamese_ipa_dataset.jsonl"):