Busy Cafe
Remove distracting cafe sounds like conversations, machines, or clattering dishes.


Before Hushv1.0

After Hushv1.0

Most Voice AI failures come from bad audio, not bad models. Hush fixes the call signal at the source, turning chaotic phone audio into clean, ASR‑ready speech so your bots, agents, and compliance teams all get better outcomes.
#5
Top‑5 speech‑enhancement model on Hugging Face’s Audio‑to‑Audio leaderboard at launch.
10,000+
Hours of real‑world noisy audio used to train Hush, including overlapping speakers and tough environments.
0.9 ms
Typical processing time for each 10 ms audio frame on CPU, keeping calls real‑time.
8 MB
Approximate model size, light enough to deploy easily in your own cloud or data center.
Make every call easier
to understand

Hush v1.0
doing on your live calls
Sudden sounds don’t break the flow
Honks, bangs, or announcements are softened so your agent can keep listening without losing words.
ASR understands more
Cleaner audio helps ASR pick up what’s being said the first time, reducing “Sorry, can you repeat that?”

























