Indonesian speech dataset

1. Product Overview

An Indonesian speech database totaling 100 hours. Sampling rate: 16,000 Hz; bit depth: 16 bits. The speakers are native Indonesian speakers aged between 7 and 70 years old. Transcriptions have been performed by professional linguists, followed by rigorous quality assurance, achieving a text accuracy rate of over 95%.

2. Text Design

Text Domains

To accommodate real-world applications, the text covers a broad range of domains, including:

• General scenarios: everyday conversations, family interactions, news broadcasts, voice-controlled commands, human-computer interaction.
• Commercial scenarios: e-commerce, catering, finance, logistics, meetings.
• Industrial scenarios: agriculture, electronics manufacturing, automotive, textiles.
• Public service scenarios: government consultations, hospital consultations, school teaching.
• Complex noise scenarios: street scenes, buses, subways, conference rooms, indoor home environments.

Considering the wide age range of speakers and their diverse application contexts, text domains are differentiated for each speaker group as follows:

• Children: Primarily focused on command-based interactions, supplemented by simple conversational texts, covering home, in-car, human-computer interaction, and general daily dialogue.
• Young and middle-aged adults: This group has the highest usage rates and the broadest coverage, encompassing home, in-car, human-computer interaction commands, dining, finance, logistics, medical consultations, industrial manufacturing, education, and casual daily chats.
• Seniors: Their domain overlaps with both children and young/middle-aged adults, covering home, in-car, human-computer interaction commands, dining, news, medical consultations/healthcare, entertainment, travel, and everyday casual conversation. For simple command phrases, we distinguish between fast, medium, and slow speaking speeds.

The specific number of sentences assigned to each speaker group is as follows:

Domain · Children · Young & Middle-Aged Adults · Seniors

Home Commands · 25*3 · 5*3 · 20*3

In-Car Commands · 25*3 · 5*3 ·

Human-Computer Interaction Commands · 25*3 · 5*3 · 20*3

Family Conversations · 50 · ·

Dining · · 100 · 55

Finance · · 90 ·

Logistics · · 100 ·

Medical Consultations/Healthcare · · 100 · 55

News · · · 55

Entertainment · · · 55

Travel · · · 55

E-Commerce · · 100 ·

Industrial Manufacturing · · 80 ·

Education and Teaching · · 80 ·

General Daily Texts · 50 · 105 · 55

Summary · 325 · 800 · 450

In addition to reading aloud, children’s recordings include guided dialogues. The thematic domains covered per speaker are as follows:

Domain · Example · Hours

Family Conversation (Parent-Child) · Parent: “What’s your favorite game? Do you enjoy playing it?” Child: “I love building with blocks—when I finish constructing a castle, I feel so proud!” · 0.1

Dialogue · ·

Educational Interaction (Teacher-Student) · Teacher: “How did you approach this math problem? Is there anything you don’t understand?” Child: “I can do addition, but I always mix up subtraction. Could you explain it again, please?” · 0.1

Casual Chat (Classmates, Friends, etc.) · A: “That game today was really fun—let’s play together tomorrow!” B: “I might not be able to make it tomorrow, but how about the day after?” · 0.05

Transcription Guidelines

Transcribers produce written content based on the audio they hear, striving to ensure that the text accurately reflects the spoken words. General guidelines include:

1. The transcribed text must exactly match the spoken audio—no extra words, missing words, or errors.
2. Oral symbols should be retained using half-width characters, with one space on either side—for example, “iskon 25% dari harga asli.” The percent sign is attached directly to the number without any spaces.
3. Proper English abbreviations should retain their full original spelling, such as USB or Wi‑Fi; continuous letter sequences require spaces between letters, e.g., “A B C”.
4. Phonetic slashes should be separated from the text by a single space, applicable only to pronunciation teaching materials—for example, “Bunyi /e/pada kata mereka berbeda”.

3. Directory Structure and Naming Conventions

Directory Structure

The data directory tree is structured as follows:

└─DOC (Documentation Files)
││├─content.txt (List of Transcribed Content)
││├─spk-info.txt (Speaker Information Table)
└─SPEECHDATA (Audio Folder)
││├─I0001 (Speaker ID)
││ │├─I0001W0001.wav (Audio File)

Naming Conventions

CORPUS/USAGE/SPEAKER_ID/SPEECH_ID, e.g., SPEECHDATA/I0001/I0001W0001.wav

Directory Name · Content · Notes

CORPUS · · Speech Database Number

USAGE · SPEECHDATA · Audio Folder

SPEAKER_ID · I0001 · Folder for Speaker ID

SPEECH_ID · I0001W0001.wav · Audio File

4. Speaker Information

Gender Information

The database contains a total of 123 speakers, comprising 60 males and 63 females, as shown below:

Gender · Male · Female · Total

Proportion · 48.8% · 51.2% · 100%

Age Information

The speakers’ ages span from 7 to 70 years old, distributed as follows:

Age Group · Number of Speakers · Percentage (%)

Ages 7–18 · 10 · 8.1%

Ages 18–35 · 105 · 85.4%

Ages 36–60 · 6 · 4.9%

Ages 61–70 · 2 · 1.6%

Total · 123 · 100%

5. Recording Plan

Recording Environment

Recordings were conducted in homes or other quiet indoor settings, free from significant background noise.

Recording Method

Speakers used self-developed recording tools for close-range recording. Recording formats included reading aloud, natural speech, and conversational dialogues:

• Reading Aloud: Reading pre-assigned texts.
• Natural Speech: Given a scenario and text, speakers produced sentences naturally and fluently without strict adherence to the provided text.
• Conversational Dialogue: Guided discussions.

Data Sample Download

Click to download a sample dataset