Select Page

  • What is is a platform to allow anyone to add accurate and fast speech recognition system to anything interactive and voice-based. is designed for children under the age of 16 and includes large-scale speech recognition models combined with powerful natural language understanding.

  • What is Speech Recognition?

    Speech recognition is a sub-field of computational linguistics that develops methodologies and technologies that enables the recognition and translation of spoken language into text by computers. It is also known as automatic speech recognition (ASR), computer speech recognition or speech to text (STT).

  • What is Edge Speech Recognition?

    Edge speech recognition describes the geographic distribution of data processing. As opposed to cloud-computing, edge-computing is done closer to the source of the device where the data is captured instead of relaying that data to a distant cloud server. With Edge Speech Recognition or Edge Speech to Text, no data ever leaves the device and all decoding takes places locally. This process is fastest and most secure way of decoding speech.

  • What is Embedded (offline)

    Embedded is an offline speech recognition that resides locally on the device without any connection to the cloud or servers. This technology is designed and built specifically to avoid collecting any data from children.

  • What is Edge

    Edge is our most advanced optimized and efficient speech to text systems  for embedded platforms. Edge brings speech recognition and AI closer to the chipset level allowing devices to perform at the edge.

  • Does have NLP and NLU?

    Yes, Kidsense has several options of standalone speech recognition, keyword spotting, wake words, NLP, and NLU.  NLP (natural language processing) and NLU (natural language understanding) can extract intents and entities from this text to allow users to perform an action such as conversation, education and command and control, while keyword spotting can be used to interrupt a conversation or identify key actions.

  • Is Secure?

    Yes, is the most secure speech recognition engine that exists for children. Both platforms (cloud-based and embedded) have extreme security measures in them to eliminate any chance of data-breach. Additionally, Embedded is completely unable to collect, store and share kids audio files making it the most secure and safe speech recognition for children.

  • Does Embedded Collect Kids Data?

    Embedded is built in such a way that is unable to collect, store, share kids audio files. This platform is uniquely designed with kids’ data privacy in mind providing a piece of mind to developers and parents.

  • Is COPPA and GDPR Compliant?

    Yes, both cloud-based and Embedded are COPPA and GDPR compliant.

  • What is the biggest advantage of over Google, Microsoft, IBM and other speech recognition systems?

    There are many advantages that offers against other ASR providers:

    1. Focus on Kids
    2. Faster Latency
    3. Better Accuracy for Kids
    4. Embedded Solutions

  • Is scalable?

    Yes is built to scale. The platform is able to support both cloud-based and embedded applications of any size in North America and APAC.

  • Who are your current clients? is currently being used by many robotics, toy developers and ai-speaker manufacturers in USA, China, and Korea. Additionally, we support many educational software programs with their language assessment needs.

  • What applications does Support?

    The platform supports hardware applications such as robotics, toys, ai-speakers, augmented reality (AR), virtual reality (VR), etc. It also supports educational software and gaming applications.

  • How many different models does have? offers various models suiting all client use, ranging from several keyword command and control phrases for non-OS microcontrollers to 1 billion word n-grams models for linux based servers. can be thought of as several parts that can all be used independently. Low powered wake word can be used to start where automatic speech recognition can translate spoken utterances into text.

  • How fast is offers different platforms and solutions with latencies ranging from 100ms to 1s.

  • What operating systems (OS) does support?

    Linux, Windows, iOS, Android, Mac, RTOS and non-os based microcontrollers

  • What languages does support?

    The speech recognition platform supports English, Mandarin, Korean Spanish and  French. Additionally German, Italian and Japanese will be added to the platform in the future.

  • Where is available? is available all around the world.

  • What are system requirements for cloud-based

    Cloud based requirements are very minimum, your system would just need to be able to send and receive data to/from the cloud.

  • What are system requirements for embedded

    We have deployed Kidsense on microcontrollers, androids, iOS, and servers and can adapt our models for your needs. Please contact us for details as each embedded system is different.

  • Does have specific domain expertise?

    Yes, both cloud-based and embedded solutions are built on kids entertainment and educational domains.

  • Does support people with special needs?

    Yes we offer customized solutions for children with special needs such as dyslexia, autism, communication disorder and speech impairment.

  • How do I get a demo without committing to buy?

    You can contact us at . We will happily set you up with a customized demo without any commitment needed.

  • How do I get started?

    Developers can get up and running within minutes. It’s also platform agnostic, so you can build it once and deploy it everywhere. Contact us and we will get you started on the same day!

  • Which SDKs are available for Houndify?

    While we’re working to implement new SDKs, here’s a list of what’s currently available:

    • Android
    • iOS
    • C++
    • Web
    • Python
    • Java
    • C Sharp
  • Have other questions?

    Contact us with your question and a member of our support team will be in touch with you shortly!