File size: 1,358 Bytes
4b53c52
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
# DataCatalyst Multilingual TTS Sample  
**Hindi | English | Hinglish (Code-Switched)**  
Version: v1.2 | April 2026  

---

## Overview

This dataset is a multilingual speech sample curated by DataCatalyst for ASR and TTS evaluation.

Languages included:
- Hindi  
- English (Indian accent)  
- Hinglish (code-switched)  

Each language contains 6 utterances (~10 seconds each).

---

## Key Features

- 48 kHz, 24-bit mono WAV  
- Speaker-consistent recordings  
- Loudness-normalized variants (EBU R128 aligned)  
- Rich annotations (emotion, style, intent, delivery attributes)  
- Explicit contributor consent  

---

## Dataset Structure

/hindi/
  /audio/
  /audio_processed/
  speaker_metadata.json
  utterances.json

/english/
  /audio/
  /audio_processed/
  speaker_metadata.json
  utterances.json

/hinglish/
  /audio/
  /audio_processed/
  speaker_metadata.json
  utterances.json

---

## Audio Specifications

- Format: WAV (PCM)  
- Sample Rate: 48,000 Hz  
- Bit Depth: 24-bit  
- Channels: Mono  
- Loudness (processed): ~ -22 to -18 LUFS  
- True Peak: ~ -6 to -3 dBTP  

---

## License

This dataset is released under the **DataCatalyst Evaluation License (DEL-1.0)**.

Permitted:
- Internal evaluation and benchmarking  

Not permitted:
- Commercial use  
- Production model training  
- Redistribution  

Contact: divyam@datacatalyst.in