AudioChinese dialects

Shanghai Dialect Speech Database

About this dataset

1. Product Overview

The Shanghai Dialect Speech Database comprises a total of 340 hours of audio. The data includes both read speech and conversational speech, with 174 hours of read speech and 166 hours of conversational speech. The recording language is the Shanghai dialect, and the recordings were made in Shanghai, China. A total of 724 speakers were invited to record in quiet environments using their mobile phones. This database has undergone professional proofreading, achieving a text accuracy rate exceeding 95%, making it suitable for research in voice recognition, speech recognition, and related fields.

2. Recording Corpus

Recording Text Results

In the read-speech portion, each speaker recorded 470 sentences, covering wake-up words, command phrases, free-form text, and more. The specific distribution is as follows:

Domain · Example Sentences · ID Numbers

Wake-Up Words · Xiao Ai Tongxue, Xiao Du Xiao Du · 0001–0021

POI · Navigate to Science and Technology Square · 0022–0044

Home Automation · Please turn on video two · 0045–0051

Phonetic Spelling · Q M P E · 0052–0054

Numbers · Eight hundred sixty zero nine thousand five hundred seventy-seven · 0055–0074

Q&A · What movies did Chen Shisan star in? · 0075–0089

Music · Sing a song by Wang Sutong · 0090–0119

TV & Movies · The Female Kingdom (Chinese version) · 0120–0139

Radio · I want to tune AM to 915.5 MHz · 0140–0144

Entertainment · Ouyang Nana later worked in mainland China · 0145–0160

Finance · Year-on-year growth of 1.9% · 0161–0240

Technology · Total shipments fell short of Apple Watch’s first-day pre-orders · 0241–0290

Sports · Going forward, we will base ourselves on North马 · 0291–0340

Current Affairs News · A notice was posted on the glass door · 0341–0390

General Topics · For you, tonight must be very long · 0391–0470

Total · · 470

During conversations, speakers freely improvised based on given topics. The topic categories are roughly as follows:

Domain · Proportion

E-commerce & Retail · 12%

Finance · 14%

Education · 13%

Technology · 13%

Life Services · 13%

Automotive Sales · 7%

Real Estate · 9%

Logistics · 6%

Other · 15%

Corpus Processing

 De-identification: Remove politically sensitive, personally private, or pornographic/violent content.  Delete symbols such as <, >, [, ], ~, /, \, =, except for numeric strings.  Eliminate content containing languages other than Chinese and English.  Standardize formatting.

3. Speaker Information

Gender Ratio

The database contains a total of 724 speakers—232 males and 492 females.

Gender · Male · Female · Total

Proportion · 32% · 68% · 100%

Age Distribution

The age distribution of speakers is as follows: A (18–25 years old); B (26–40 years old); C (41–50 years old); D (51+ years old). Age group proportions are shown below:

A · 18–25 years old · 79 · 10.9%

B · 26–40 years old · 374 · 51.7%

C · 41–50 years old · 223 · 30.8%

D · 51+ years old · 48 · 6.6%

Total · 724 · 100% ·

4. Data Recording Environment

Recording Environment

A quiet indoor setting, free from echoes and extraneous noise.Recording Equipment

Common smartphones such as Apple, Huawei, Xiaomi, OPPO, etc. Data is stored in 16 kHz, 16-bit WAV format.Recording Method

For read-speech segments, speakers recorded via a mobile app in a quiet room, reading texts at normal speed. For conversational segments, speakers recorded phone calls in a quiet environment.

5. Speech Data Transcription

Transcribers convert audio into written text, striving to ensure consistency between spoken content and written form. General guidelines include: 1) Transcriptions must exactly match the spoken audio—no extra words, omissions, or errors. 2) Numerals should be transcribed in Chinese characters—for example, “one two three” instead of “123.” Pay attention to distinguishing “one” from “yao,” and “two” from “liang.” 3) English pronunciation in audio should be rendered in corresponding Chinese characters or English. Specific cases: All letters or words in URLs should be capitalized—for example, “www.abc.com” becomes “Three W point A B C point com.” English words within the audio should be transcribed entirely in lowercase; English letters should be capitalized. 4) Annotation completeness must align with actual pronunciation—no deletions allowed.

6. Data File Directory

Directory Structure

The directory tree structure is as follows:

└─dialogue（Conversation Data）
│├─DOC（Text Folder）
│├─wav_list.txt（Audio List）
│├─spkrinfo.xlsx（Speaker Information）
│├─Textgrid（Transcription Text Folder）
│├─PHD_0006.Textgrid（Transcription Text）
│├─SPEECHDATA（Data Folder）
││ ││ PHD_0006.wav（Audio File）
│
└─reading（Read-Speech Data）
│├─DOC (Text Folder)
│├─content.txt（Transcription Text）
│├─wav_list.xlsx（Audio List）
│├─spkrinfo.xlsx（Speaker Information）
│├─SPEECHDATA (Data Folder)
│├─PH0012 (Speaker ID)
││ ││ PH0012_0001.wav（Audio File）