DNSMOS gets you most of the way there. It doesn’t tell you if the data is right for your model.

Written by DJ Jalali, Chief Technology Officer, with Geoff Bremner (Senior Audio Software Engineer, AI & Datasets) and Justin Hugh (Senior Product Manager, Voice Data)

Fill out the form below to get the steps needed to
maximize your source data.

Marketing opt-in

Trusted by the world’s leading enterprises, AI technology, and software companies

Microsoft Logo SoundHound Logo Cresta Logo BMW Logo

Learn more about:

  • 1

    The limitations of DNSMOS 

  • 2

    A four-step calibration framework you can apply to your own model

  • 3

    How Voices validates audio quality in producing custom datasets

Why ML teams read this

You’re building or fine-tuning a speech model and you’ve hit the same wall: DNSMOS tells you a score, not whether the data is right for your model. If you’re not calibrating DNSMOS and setting your own thresholds for acceptable data quality, you’re missing out on usable source data—increasing your training time and costs. This article will explain how to calibrate DNSMOS to your specific use case and maximize your source data.

What’s inside

Unsubscribe anytime. We’re not going to spam you.

Unsubscribe anytime. We’re not going to spam you.