haoxiangsnr/llm-tse
Typing to Listen at the Cocktail Party: Text-Guided Target Speaker Extraction (LLM-TSE)
This helps you isolate a specific speaker's voice from a noisy audio recording, much like how people focus on one conversation in a crowded room. You provide an audio recording with multiple speakers and a text description of the speaker you want to hear. The output is a clear audio track of only that speaker's voice. This is for anyone who needs to extract specific voices from complex sound environments.
No commits in the last 6 months.
Use this if you need to cleanly separate one person's voice from a chaotic audio recording using only a text description, without needing a prior voice sample of that person.
Not ideal if you already have a high-quality, pre-recorded voice sample (voiceprint) of the target speaker you wish to extract.
Stars
42
Forks
2
Language
JavaScript
License
—
Category
Last pushed
Oct 13, 2023
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/llm-tools/haoxiangsnr/llm-tse"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
little51/llm-dev
《大模型项目实战：多领域智能应用开发》配套资源
Ahmet-Dedeler/ai-llm-comparison
A website where you can compare every AI Model ✨
Michaelgathara/llm-timeline
Visualize LLM Progress Overtime
nicucalcea/sheets-llm
Use Large Language Models (LLM) in Google Sheets
cohere-ai/sandbox-grounded-qa
A sandbox repo for grounded question answering with Cohere and Google Search