Skip to content

Dataset · BAAI

Mandarin elderly conversational speech corpus (75-85 yrs)

BAAI/SeniorTalk

Audio recordings of elderly native Mandarin speakers with annotations for ASR and speaker diarization.

What's inside

An open speech dataset by BAAI in Mandarin Chinese, including 10 to 100 thousand audio and text entries. It features conversations of elderly individuals aged 75 to 85 with detailed annotations. The corpus was created to address the shortage of speech data for this age group.

What you can do with it

  • Adapting Automatic Speech Recognition (ASR) systems for elderly voices
  • Training speaker verification and diarization models (SV, SD)
  • Research in speech editing and voice assistants for older generations
Tasks
Automatic speech recognition
Languages
Chinese
Size
10K–100K rows
Downloads this month
2.3K
Likes
45

How to load it

from datasets import load_dataset

ds = load_dataset("BAAI/SeniorTalk")