kehanlu/DeSTA2
Code and model for ICASSP 2025 Paper "Developing Instruction-Following Speech Language Model Without Speech Instruction-Tuning Data"
This project offers a way for researchers and AI developers to create advanced speech language models (SLMs). It takes raw audio inputs and plain text instructions, then outputs intelligent, conversational responses about the audio content. This tool is for those building AI systems that can understand and respond to spoken language.
123 stars. No commits in the last 6 months.
Use this if you need to develop an AI model that can understand complex spoken commands and provide detailed, intelligent responses about audio content, without needing extensive speech instruction data.
Not ideal if you are an end-user simply looking for a ready-to-use voice assistant; this is a development tool for building such systems.
Stars
123
Forks
10
Language
HTML
License
—
Category
Last pushed
Jul 15, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/transformers/kehanlu/DeSTA2"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
DaoD/INTERS
This is the repository for our paper "INTERS: Unlocking the Power of Large Language Models in...
declare-lab/instruct-eval
This repository contains code to quantitatively evaluate instruction-tuned models such as Alpaca...
Haiyang-W/TokenFormer
[ICLR2025 Spotlightš„] Official Implementation of TokenFormer: Rethinking Transformer Scaling...
hkust-nlp/deita
Deita: Data-Efficient Instruction Tuning for Alignment [ICLR2024]
TIGER-AI-Lab/VisualWebInstruct
The official repo for "VisualWebInstruct: Scaling up Multimodal Instruction Data through Web...