KugelAudio 发布 Clarity-1：面向语音智能体的实时语音增强模型

Best Products Orbit Awards Awards powered by what reviewers actually sayAI Workflow Automation →AI Dictation Apps → Trending Categories Engineering & Development LLMs Productivity Marketing & Sales Design & Creative Social & Community Finance AI Agents Other Categories Trending CategoriesVibe Coding Tools AI Dictation Apps AI notetakers Code Review Tools No-code Platforms Figma Plugins Static site generators Engineering & DevelopmentVibe Coding Tools AI Coding Agents AI Code Editors LLMsAI Chatbots AI Infrastructure Tools Prompt Engineering Tools ProductivityAI notetakers Note and writing apps Team collaboration software Search AI Workflow Automation Marketing & SalesLead generation software Marketing automation platforms Design & CreativeVideo editing Design resources Graphic design tools AI Generative Media Social & CommunitySocial Networking Professional networking platforms Community management FinanceAccounting software Fundraising resources Investing AI AgentsAI Voice Agents
Launches Launch archive Most-loved launches by the community Launch Guide Checklists and pro tips for launching
News Newsletter The best of Product Hunt, every day Stories Tech news, interviews, and tips from makers Changelog New Product Hunt features and releases
Forums Forums Ask questions, find support, and connect Kitty Points Leaderboard The highest scoring community members Streaks The most active community members Events Meet others online and in-person
Advertise

SubscribeSign in

Image 1: KugelAudio

KugelAudio

Real-time text-to-speech model you can self-host

159 followers

Real-time text-to-speech model you can self-host

159 followers

Visit website

Text-to-Speech Software

•

Realtime Voice AI

Most natural real-time TTS with voice cloning and sub-60ms latency, on-prem or via API. Grammar-aware normalization reads phone numbers, IBANs, addresses, and medications naturally across 25+ languages, with word-level timestamps and IPA support. Adapters for LiveKit, Pipecat, and Vapi. Built by 4 in Berlin.

This is the 2nd launch from KugelAudio. View more

Image 2: Clarity

Clarity

Launched this week

Mute the room, keep the speaker, in real time

Visit

Real-time speech enhancement and target speaker extraction for voice agents. Clarity-1 strips out background noise and other people talking on a live call, so your agent hears only the caller. Streams as audio arrives. Hear the before/after on our site.

Image 3: Clarity gallery image

Image 4: Clarity gallery image

Image 5: Clarity gallery image

Image 6: Clarity gallery image

Free

Launch tags:Artificial Intelligence•Tech•Audio

Launch Team

Image 9: Eney

Eney Proactive AI for Mac. Free test, 20,000 credits.

Try Eney

Promoted

What do you think? …

Kajo Kratzenstein

KugelAudio

Hunter

📌

Hey everyone, we're back, this time out of Berlin! Our customers keep asking us for the same thing: make agents work when the caller isn't somewhere quiet. Voice agents work well in quiet rooms. Real callers aren't in quiet rooms. They're in cafés, on busy streets, commuting, often with someone talking right next to them. The agent hears all of it, so the caller gets misheard, or the agent gets interrupted by sounds that were never meant for it. We couldn't find a denoiser we were happy shipping to our customers, so we built one. Clarity-1 does two things on a live audio stream, in real time: Speech enhancement: removes background noise like traffic, trains and café chatter Target speaker extraction: removes other voices too, so your agent only hears the person it's talking to It processes audio as it arrives, so it slots into live calls. On the benchmarks we ran (DNSMOS), Clarity-1 had the highest overall quality score of the models we compared, for both noise removal and target-speaker isolation. The full table and before/after samples are here: https://www.kugelaudio.com/en/cl... Try it on your own audio in the dashboard or via the API: https://www.kugelaudio.com/en/si.... Your first month is completely free. Two things we'd love from you: What's the noisiest place your users call from? If you find audio where Clarity breaks, send it our way. That's the most useful feedback we can get. This is the first of several models we're releasing over the next few weeks. More soon.

Upvote (4)

Report

2d ago

Priya K

@kajo_krdoes this still work reliably if the caller is using a really cheap phone microphone?

Upvote (2)

Report

1d ago

Eliezer Cernin

@kajo_kr @priya_kushwaha1Good question.

Upvote (2)

Report

1d ago

Kajo Kratzenstein

KugelAudio

Hunter

hey @priya_kushwaha1, do you mind sharing a sample with me?

Upvote (1)

Report

1d ago

Alexander Netz

KugelAudio

Maker

@kajo_kr @priya_kushwaha1The model learned to upsample from 8khz to 24khz. Bad mic quality is part of the model training.

Upvote (2)

Report

1d ago

Gal Dayan

Dial

Likely AI

target speaker extraction on a live call is a much harder problem than noise suppression - does Clarity-1 need a short enrollment clip of the caller's voice to lock onto them, or does it figure out "the one talking to the agent" cold on a brand new call with zero prior samples?

Upvote

Report

9h ago