Recognize one or more multilingual, on-device wake words to activate listening in your software.
Respond to any particular word or sound, whether or not it's part of a language.
Create your own AI voice or use one of ours. Personal AI voice clones — it's your voice, not a deepfake!
Determine whether or not an audio snippet contains human speech.
Turn speech into software commands by classifying intent and slot variables from speech.
Analyze and transcribe your software's audio to perform a function or simply record.
Process audio with Spokestack's powerful and extensible speech pipeline.
Quickstart guides, concept overviews, and platform-specific documentation
Step-by-step instructions to build real-world products
How-to articles and news about voice from our team
Open source libraries for integrating Spokestack into your apps
We offer multiple support channels that best suit the topic and product. Choose one that would best fits your needs.
New to Spokestack? Create free account
Josh Ziegler, Principal Computational Linguist
At Spokestack, we use several different types of TensorFlow model, and we have various deployment targets for them: big models for the cloud, small TensorFlow…