17 categories and 75 products and companies, charted by an ai agent.
On-device ASR engines are speech recognition systems that run locally on devices without cloud connectivity, enabling private and low-latency voice input. Used by smartphone manufacturers, IoT device makers, and developers building privacy-focused voice applications.
whisper.cpp · faster-whisper · vosk · speechmatics on-device · nvidia riva · alphacep vosk · kaldi
On-device text-to-speech technology that converts written text into spoken audio locally without cloud dependency. Used by accessibility software developers, mobile app makers, automotive manufacturers, and smart device creators.
coqui tts · piper · kokoro · sherpa-onnx tts · mimic 3 · elevenlabs · cartesia · coqui · sesame
On-device voice assistants that process speech locally without cloud connectivity, enabling privacy-focused voice control and natural language interaction. Used by smartphone and device manufacturers prioritizing user privacy and offline functionality.
apple siri on-device · google assistant on-device · amazon alexa · samsung bixby · meta ai voice · xiaomi xiao ai
picovoice porcupine · openwake word · snowboy · sensory trulyhandsfree
On-device voice NLU processes natural language understanding locally on smartphones, wearables, and IoT devices without cloud connectivity, enabling privacy-first voice assistants and commands. Used by developers building voice features and end-users seeking offline voice control.
picovoice rhino · snips nlu · rasa · wit.ai on-device
whisper hearing · apple airpods on-device stt · nuance hearing
On-device speech translation converts spoken words from one language to another directly on a user's device without cloud connectivity, enabling real-time multilingual communication for travelers, business professionals, and international teams.
seamlessm4t on-device · apple translate app · google interpreter mode · fleurs lite
On-device voice systems that provide ambient clinical documentation and real-time speech recognition without cloud transmission, used by healthcare providers for hands-free note-taking and workflow efficiency.
suki · nuance dax · abridge · nabla copilot · sully ai
qualcomm ai voice sdk · arm ethos npu · syntiant ndp · ambiq apollo
Software models that perform speech-to-text conversion entirely on a device without cloud connectivity, used by developers building mobile apps, embedded systems, and offline voice applications.
whisper · nvidia parakeet · moonshine · sherpa-onnx · picovoice
Edge voice SDKs and runtimes are software libraries and execution environments that run voice processing on devices without cloud dependency. Developers building mobile apps, embedded systems, and IoT devices use them to add speech capabilities locally.
qualcomm ai hub · apple speech framework · deepgram · argmax · cactus
Private dictation apps capture voice input on a user's device and convert it to text without sending audio to external servers. Professionals, students, and knowledge workers use these apps to record notes, memos, and documents hands-free while keeping recordings private.
superwhisper · macwhisper · wispr flow · voiceink · aqua voice · dragon
Software voice assistants that run entirely on a device without cloud connectivity, responding to voice commands for device control and information retrieval. Used by smartphone, tablet, and smart home device manufacturers and their users.
siri · google assistant · gemini nano · alexa · bixby
Software that transcribes spoken audio to text and generates synchronized captions in real-time on a device without sending data to servers. Used by content creators, broadcasters, accessibility professionals, and meeting participants who need live captions.
live captions · otter · notta · transcribe
On-device voice speech processing that runs speech recognition, synthesis, or understanding directly on consumer hardware without cloud dependency. Developers, device manufacturers, and end-users of phones, watches, earbuds, and embedded systems use this to process voice commands and audio locally.
cerence · soundhound · sensory · dsp group
On-device edge speech models are machine learning models that process voice audio locally on user devices without sending data to cloud servers. Mobile app developers, voice assistant makers, and hearing aid manufacturers use them to build speech recognition, synthesis, and understanding features that work offline.
On-device voice speech AI processes spoken audio locally on user hardware without cloud transmission, used by developers and consumers in phones, wearables, and embedded systems. The systems perform tasks like recognition, understanding, synthesis, or translation entirely on the device.