Smart robot voice data
About this dataset
1. Product Overview
The intelligent robot voice database contains a total of 23 hours of recordings. The recording language is Chinese, and the recordings were made in China. The audio transcripts consist of intelligently controlled robot utterances, with each speaker providing 11 sentences. A total of 646 speakers were recruited, primarily from northern China, to participate in the recording process. The recordings were conducted in real home environments, simulating the use of smart home appliances and applications, with seven recording positions set up. Five of these positions used high-fidelity microphones (44.1 kHz, 16-bit), while one was recorded using an iOS smartphone (16 kHz, 16-bit) and another using an Android smartphone (16 kHz, 16-bit). This database has been transcribed and annotated by professional speech annotators and rigorously quality-checked, achieving a text accuracy rate exceeding 95%.
2. Product Catalog Structure
Catalog Structure
Data catalog structure data catalog structure
└─DOC（Text Description File）
││├─wav_list.txt（Audio List）
││├─content.txt（Speech Text Content）
││├─spk_info.xlsx（Speaker Information）
││├─room_info.xlsx（Scene Information）
│
└─SPEECHDATA（Data Folder）
│├─0001（Speaker Folder）
││ ││ H0001A（Audio Folder）
││ ││ H0001A0001.wav（Audio File）
││ ││ H0001B（Audio Folder）
││ ││ H0001B0001.wav（Audio File）
Naming Rules
CORPUS/USAGE/SPEAKER_NUM/POINT_ID/SPEECH_IDe.g./SPEECHDATA/0001/H0001A/ H0001A0001.wav
Directory Name · Content · Remarks
CORPUS · · Voice Database Number
USAGE · SPEECHDATA · Folder Name
SPEAKER_NUM · 0001 · Speaker Folder Name
POINT_ID · H0001A · Recording Position Number
SPEECH_ID · H0001A0001.wav · WAV File
3. Text Design
Corpus Creation
Considering the application of speech recognition in mainstream smart home devices, the text design adopts robot-related content. The corpus pool is processed according to established rules: ⚫ De-identification: Politically sensitive, personally private, or pornographic/violent content is removed. ⚫ Removal of symbols such as <, >, [, ], ~, /, \, =. ⚫ Exclusion of content containing languages other than Chinese and English. ⚫ Elimination of single sentences exceeding 25 characters. ⚫ Standardization of format.
Text Structure
Serial Number · Domain · Allocated Quantity per Entry/Sentence
1 · Robotics · 11
Total · · 11 sentences
4. Recording Environment
Recording Site
1. The recording scene replicates a real home environment, including basic household items, appliances, tables, and chairs. 2. The specific recording position on-site is determined based on the actual environment (including distance and height considerations). 3. Recorded distances vary between 0.5 m, 1.5 m, 2 m, 2.5 m, 3 m, 3.5 m, 4 m, and 5 m, depending on the situation. 4. After finalizing the recording location and positions, equipment is calibrated, and preparatory work—including actual recording—is carried out to ensure that all recording devices function properly, remain stable, and meet recording requirements. 5. Speakers record their voice data at designated locations.
Recording Equipment
Recording equipment includes high-fidelity microphones, tape recorders, and smartphones. This database stores data in two formats: high-fidelity recordings at 44.1 kHz, 16-bit mono, and smartphone-recorded data at 16 kHz, 16-bit mono.
Recording Method
In a home setting, seven recording positions are arranged—three close-range positions and four functional positions. Functional positions are fixed near air conditioners, televisions, table lamps, and robotic vacuum cleaners; close-range positions are located approximately 30 cm from the speaker, equipped with one high-fidelity microphone and two smartphones (iOS and Android). Speakers deliver their recordings at normal volume and pace, reading aloud the provided text.
Serial Number · Content
A · Recording Position 1: Air Conditioner
B · Recording Position 2: Television
C · Recording Position 3: Table Lamp
D · Recording Position 4: Robotic Vacuum Cleaner & Others
E · Recording Position 5: Close-Range iOS
F · Recording Position 6: Close-Range Android
G · Recording Position 7: Close-Range Microphone
5. Annotation and Transcription Guidelines
Transcribers render the spoken content into written form, striving to maintain consistency between the text and the original audio pronunciation. Key principles include: 1) Transcriptions must exactly match the spoken words—no extra words, omissions, or errors. 2) Numbers should be written in Chinese characters, such as “一二三,” rather than “123.” Pay special attention to distinguishing “一” from “幺,” and “二” from “两.” 3) English words appearing in sentences should be transcribed phonetically—for example, “thank you”—and letters pronounced individually should be rendered in uppercase followed by spaces, such as “N B A” or “U F O.” 4) Symbols within sentences should be transcribed according to the speaker’s actual pronunciation—for instance, “三W点百度点com.” Unpronounced symbols should be omitted. Brand names and proper nouns should follow customary usage, such as “QQ空间,” “iPhone,” or “喜马拉雅.” 5) The completeness of annotations must align with the actual pronunciation; no deletions are permitted.
6. Speaker Information
Basic Information Recording
Speaker information includes task number, gender, accent region, age range, and environmental context.
Task Number · Gender · Accent Region · Age Range · Environmental Context
0001 · Male · Northern · B · Scene One
Task Number: Each speaker receives one task number, corresponding to one recording text. Each speaker may only participate once. Gender: Male or female. Accent Region: Classified according to the speaker’s native regional dialect—Northern, Southern, or Other. Age Range: A (under 15), B (16–25), C (26–40), D (41+). Environmental Context: Twelve recording scenes are defined, with scene details listed in the attached room_info file.
Scene · Number · Scene · Number
Scene One · bedroom_1 · Scene Seven · diningroom_1
Scene Two · bedroom_2 · Scene Eight · drawingroom_1
Scene Three · bedroom_3 · Scene Nine · drawingroom_2
Scene Four · bedroom_4 · Scene Ten · drawingroom_3
Scene Five · bedroom_5 · Scene Eleven · drawingroom_4
Scene Six · bedroom_6 · Scene Twelve · drawingroom_5
Speaker Demographics
Gender Ratio
The database comprises a total of 646 speakers—276 males and 370 females.
Gender · Male · Female · Total
Ratio · 43% · 57% · 100%
Age Distribution
A (under 15): 0; B (16–25): 631; C (26–40): 12; D (41+): 3.
Age Group · Age Bracket · Number of Speakers · Proportion · Male · Female
A · ＜15 years old · 0 · 0 · 0 · 0
B · 16–25 years old · 631 · 97.7% · 269 · 362
C · 26–40 years old · 12 · 1.9% · 6 · 6
D · ＞41 years old · 3 · 0.5% · 1 · 2
Total · · 646 · 100% · 276 · 370
Dialect Distribution
Northern: 545; Southern: 91; Other: 10.
Region · Number of Speakers · Proportion
Northern · 545 · 84.4%
Southern · 91 · 14.1%
Other · 10 · 1.5%
Total · 646 · 100%
Sample Data Download
Samples
- Audio sample 1