versae commited on
Commit
39113db
·
verified ·
1 Parent(s): 93c7b66

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +63 -1
README.md CHANGED
@@ -11,4 +11,66 @@ tags:
11
  pretty_name: NB-LibriVox
12
  size_categories:
13
  - 1K<n<10K
14
- ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
11
  pretty_name: NB-LibriVox
12
  size_categories:
13
  - 1K<n<10K
14
+ ---
15
+
16
+ # 📄 NB-LibriVox
17
+
18
+ A high-quality Norwegian text-to-speech (TTS) dataset derived from LibriVox public domain audiobooks. It includes audio clips with pseudo-aligned transcripts and punctuation, curated by the National Library of Norway for speech synthesis and ASR research.
19
+
20
+ ---
21
+
22
+ ## 📂 Dataset Overview
23
+
24
+ | **Field** | **Description** |
25
+ |-------------------|---------------------------------------------------------------------------------|
26
+ | `file_name` | Audio file name in `.wav` format |
27
+ | `id` | Unique identifier for each utterance/sample |
28
+ | `text` | Transcript automatically generated using NB-Whisper Large and nb-wav2vec2 |
29
+ | `text_punct` | Transcript with punctuation automatically added using Qwen3 |
30
+ | `speaker_id. ` | Identifier of the speaker |
31
+ | `speaker_name` | Name of the speaker |
32
+ | `speaker_gender` | Gender of the speaker (e.g., `male`, `female`, `unknown`) |
33
+ | `speaker_status` | Speaker's native status (e.g., `L1`, `L2`) |
34
+ | `title` | Title of the source audiobook |
35
+ | `author` | Author of the original text |
36
+ | `sequence` | Position of the utterance within the audiobook file |
37
+ | `duration` | Duration of the audio clip in seconds |
38
+
39
+ ---
40
+
41
+ ## 📌 Additional Information
42
+
43
+ - **Language**: Norwegian (Bokmål)
44
+ - **Audio Format**: MPEG ADTS, layer III, v1, 64 kbps, 48 kHz, Monaural
45
+ - **Text Alignment Method**:
46
+ - **ASR Model**: [`NB-Whisper Large`](https://huggingface.co/NbAiLab/nb-whisper-large)
47
+ - **Supplementary Model**: [`nb-wav2vec2`](https://huggingface.co/NbAiLab/nb-wav2vec2-1b-bokmaal-v2)
48
+ - **Alignment Type**: Pseudo-labeling (not force-aligned)
49
+ - **Punctuation Model**: [Qwen3 235B](https://huggingface.co/Qwen/Qwen3-235B-A22B)
50
+ - **Total Duration**: ~8 hours
51
+ - **Total Samples**: 4907 samples
52
+
53
+ ---
54
+
55
+ ## 📜 Licensing & Usage
56
+
57
+ - **License**: Creative Commons Zero (CC0)
58
+ - **Data Source**: LibriVox public domain recordings
59
+ - **Dataset Creator**: National Library of Norway
60
+
61
+ > Free to use, modify, and distribute for any purpose, including commercial use.
62
+
63
+
64
+ ---
65
+
66
+ ## ⚠️ Ethical Considerations
67
+
68
+ - All content is in the public domain and CC0-licensed.
69
+ - Speakers are volunteers; synthesized voices should not be used in misleading or impersonating contexts.
70
+ - Encourage transparency when using in generative applications (e.g., voice assistants, synthetic speech).
71
+
72
+ ---
73
+
74
+ ## 📎 Citation
75
+
76
+ > “Norwegian LibriVox TTS Dataset (CC0), curated by the National Library of Norway. Transcripts pseudo-labeled using NB-Whisper Large and nb-wav2vec2.”