styfeng/TinyDialogues

Code & data for the EMNLP 2024 paper: Is Child-Directed Speech Effective Training Data for Language Models?

38
/ 100
Emerging

This project helps researchers and computational linguists explore how language models learn from speech directed at children versus adults. It provides tools to process child-directed speech and adult speech datasets, format them for training, and then train and evaluate language models like GPT-2 and RoBERTa on this data. The output helps understand the effectiveness of different types of linguistic input.

No commits in the last 6 months.

Use this if you are a computational linguist or cognitive scientist studying language acquisition and want to investigate how different speech environments impact the development of language models.

Not ideal if you are looking to train a general-purpose, production-ready language model or if you are not working with child-directed speech datasets.

computational-linguistics language-acquisition natural-language-processing cognitive-science speech-research
Stale 6m No Package No Dependents
Maintenance 2 / 25
Adoption 5 / 25
Maturity 16 / 25
Community 15 / 25

How are scores calculated?

Stars

12

Forks

5

Language

Python

License

MIT

Last pushed

Oct 04, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/transformers/styfeng/TinyDialogues"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.