LLaSE-G1 is a unified speech enhancement model capable of handling multiple tasks without extra task prompts, including:<p><pre><code> Noise Suppression (SE)
Target Speaker Extraction (TSE)
Packet Loss Concealment (PLC)
Acoustic Echo Cancellation (AEC)
Speech Separation (SS)
</code></pre>
huggingface link:<p>> <a href="https://huggingface.co/ASLP-lab/LLaSE-G1" rel="nofollow">https://huggingface.co/ASLP-lab/LLaSE-G1</a><p>demo page:<p>> <a href="https://submission-papers.github.io/LLaSE-G1-demo-page/" rel="nofollow">https://submission-papers.github.io/LLaSE-G1-demo-page/</a><p>arxiv paper:<p>> <a href="https://arxiv.org/abs/2503.00493" rel="nofollow">https://arxiv.org/abs/2503.00493</a>