Introduction

Corti foundation models are specifically designed for use in the healthcare domain. While the speech recognition models are continuously refined, a tier system has been introduced to relay functionality and performance that is available.

Selecting language models at the appropriate tier allows you to find the right fit for your needs while balancing capabilities and cost.

Corti ASR tiers

TierDescriptionPrimary use case
Base“Base” ASR models enable speech-to-text with a general medicine model for a variety of medical specialtiesBroad support for speech-enabled clinical documentation workflows
Enhanced“Enhanced” ASR models enable speech-to-text with optimized medical terminology recognition for a variety of specialties, balancing speed, functionality, and qualityPower ambient documentation workflows
Premier“Premier” ASR models enable speech-to-text with the highest accuracy, widest language model and utmost functionality (e.g., customized commands, user-defined lexicon, medical specialist terminology)Power robust dictation or ambient documentation workflows

Available languages

LanguageLanguage Code (BCP47)Tier availability
EnglishenBase, Enhanced, Premier
English (US)en-USBase, Enhanced, Premier
English (UK)en-GBBase, Enhanced, Premier
DanishdaBase, Enhanced, Premier (beta)
GermandeBase, Enhanced, Premier (beta)
Swiss-Germande-CHBase, Enhanced (beta)
FrenchfrBase, Enhanced (beta)
SwedishsvBase, Enhanced (beta)
SpanishesBase, Enhanced (coming soon)
DutchnlBase
NorwegiannoBase
ItalianitBase
PortugueseptBase
ArabicarBase (coming soon)

If you are interested in a language that is not listed here, please contact your Corti representative