On-device Speech Recognition
Software Development Kits

iOS, Android, Web, and custom platforms

English (adults, kids), Spanish, German, and French

Novel Effect

Read aloud with your favorite print books and Novel Effect's award-winning app follows along playing music, sound effects, and character voices at just the right moment.

KeenASR SDK powers on-device speech recognition in Novel Effect iOS and Android apps.


MiraCheck CoPilot is a virtual co-pilot for sport pilots. Using this app sport pilots can perform various tasks, like pre-flight/in-flight checklists, querying for information, etc. all hands-free, just by using their voice.

KeenASR SDK powers always-on listening in MiraCheck CoPilot app.


Ambifi's SaaS platform empowers everyone with the ability to create virtual “copilot” guides, smart procedures & protocols, checklists, workflow learning, interactive guidelines/playbooks and more.

KeenASR SDK powers always-on speech recognition in Ambifi iOS and Android mobile apps.

AudioKit Pro

From the creators of the AudioKit, one of the most popular open-source audio projects for iOS, comes Hey Metronome, a hands-free, voice controlled metronome app.

Hey Metronome app eliminates the problems of practicing with a classic metronome and its many uninspired digital equivalents. With Hey Metronome you don't need to put your instrument down in order to change the settings; metronome control is done hands-free, just by using your voice.

KeenASR SDK powers continuous-listening speech recognition in Hey Metronome app.


Readability is a learning app that helps improve reading and comprehension in young and struggling readers. Practicing pace and pronunciation through reading aloud helps kids build the foundation needed to become excellent readers. With Readability, they'll also learn a variety of reading skills including accuracy, fluency, speed, decoding, and comprehension

PBS Kids

Build a robot and immerse your kids in STEM learning! The Cat in the Hat Invents engages your preschooler in engineering and problem solving. Join Nick, Sally and the Cat in the Hat as they explore the world of science, working through obstacles as they overcome various engineering challenges.

KeenASR SDK powers speech recognition functionality in The Cat In The Hat Invents app.

Elm Park Labs

KeenASR SDK powers voice interactions in Elm Park Labs VR training products.

eoStar Logo

eoPick is a voice-picking technology that virtually removes the need for list-based picking and clipboards. Warehouse employees can use this picking software with a headset or Android device, so they have both hands free at all times to retrieve what they need.

Noggin Logo

KeenASR SDK powers voice interface in interactive videos in Noggin app.

Almedia Logo

Avenkraft is an iOS and Android game for learning German and French.

Play as an apprentice Superhero who will have to travel through the regions of the Upper Rhine to become hero number one!

KeenASR SDK powers on-device speech recognition in Avenkraft iOS and Android apps.

Rally Reader Logo

Rally Listens, Gives Feedback & Tracks Progress

Rally is an AI-powered app that hears you read, identifies errors, tracks accuracy on a word-by-word basis, and provides real-time feedback.

KeenASR SDK powers on-device speech recognition in Rally Reader iOS app.

University of Louisville Logo

The University of Louisville, in collaboration with a large medical device company, is exploring the use of voice controls for a medical device for physically impaired patients.

Osmo Logo

Reading Adventure by Osmo is an interactive learn-to-read program that combines innovative speech recognition with real physical books.

StudyCat Ltd. Logo

KeenASR SDK powers learning experiences in StudyCat mobile apps.

Numina Group Logo

KeenASR SDK powers speech recognition in Numina\'s RDS Victory Voice™ Picking System

Pragmatica Logo

KeenASR SDK powers language therapy activities in Pragmatica's Virtual Reality app.

Why on-device?

There is no need to rely on 3rd-party cloud services or build and scale your own ASR backend when on-device speech recognition is viable for a number of use cases

Works Offline

Works Offline

No internet connectivity is required

Privacy And Security

Privacy and Security

Speech data stays on the device; COPPA/HIPAA compliant

Cost Effective

Cost Effective

No mismatch between revenue and cost model; no usage-based fees



Language and acoustic models can be customized to specific domain


Deep Neural Networks

Deep Neural Networks

State-of-the-art engine based on Deep Neural Networks

On Device Processing

On-device Processing

All processing occurs locally; up to 100k words on most recent devices

Real Time Processing

Real-time Processing

Results are provided in real-time

Development Tools

Development Tools

Cloud based tools for development support

Nlu Dialog Management


Natural Language Understanding extracts user’s intent (coming soon)

We provide professional services for SDK integration, proof-of-concept development, customization of language and acoustic models, and porting to custom hardware platforms.

Use cases

AR and VR


Extend your virtual and augmented reality apps with voice interaction



Simplify UX in mobile apps with voice search

Voice Control in Apps

Voice control in apps

Circumvent complex navigation with voice commands


Review SDK documentation and simple proof-of-concept iOS and Android apps on Github.

Debug, evaluate, and tune your apps and gain insights about speech recognition performance via Dashboard, a cloud-based tool that works in conjuction with KeenASR SDK. Use of Dashboard is free for our customers during the app development process.

Our trial SDK works for 15min at the time, and allows you to experiment with the SDK in one of our proof-of-concept apps or in your own app.