Voice search done right

Speechly enables voice search experiences that give real-time visual feedback to user enabling them to correct themselves naturally.

Integrating Voice Search with Speechly

End-to-end example for creating voice search for web with Speechly. Available also for iOS and Android!

1 Sign up to start

Sign up to Speechly Dashboard to configure your model and to get your application ID.

You can copy and paste the example configuration from the right to get started and learn more about the configuration in our documentation.

Sign up
Example configuration

2 Add Speechly UI Components

Add customizable UI components with a few lines of code and leave the hard parts of handling audio for us.

UI components available for web, iOS and Android.

3 Integrate to your application

Our fully streaming API starts returning results from the second the user starts speaking.

Loop through the results and update the filters accordingly.

Read more in documentation
Sign up

Try out the full code

Start the experience by clicking the power button and give permissions to microphone in your browser. After that, click and hold the microphone and say something like "Show red Puma" and see the dropdown menus changing accordingly. You can see the full source code in the left.

See the Pen Speechly Voice Search Example by Hannes Heikinheimo (@hheikinh) on CodePen.

Faster and more intuitive

Voice search is several times faster and more natural way for creating search experiences, especially on mobile. It can be used in any content-heavy domain that depends on good search, such as eCommerce.

”It was an amazing experience using Speechly”


Intent and entity extraction
Our API returns user intent and entities along with the transcript. Easy to configure for any use case.
Powerful and easy-to-use Speechly Annotation Language for configuring your model.
Single configuration
Same model available for all platforms for consistent user experience. Fully streaming real-time API.
Get insight into how your users interact with your application.
Multiple intents
Supports as many intents and entities per utterance as you need.
Command line tools
Integrates easily to your existing workflows and processes.
No data needed
Get up and running with no prior datasets.
Client libraries and tutorials for easy integration.
Low latency
Real-time visual feedback for intuitive and easy user experience.
Tailored speech recognition
Supports your custom vocabulary.
Optimized for touch screens
Integrates to your current user interface.
Data security built-in
Your users' data is secured with the highest industry standards

Start building awesome voice search experiences

Submit your email address and get access to our platform

Fields marked with * are required


Available on all major touch screen platforms