Azure AI Speech vs Kardome VoiceID

Azure AI Speech has no write-up on Specifiction yet; its figures are above.

Kardome VoiceID is a paid biometric voice-recognition module for identifying speakers enrolled in advance and adding voice authentication to applications. It verifies identity through voice characteristics such as pitch and pronunciation, allowing access control for devices or data. VoiceID sits within Kardome’s Cognition AI technology, which combines speaker identity with intent and context. Related modules include Wake Detection for activation through wake words, context, or zonal presence; SoundSense for detecting and identifying ambient sound; and InferLM, an on-device small-model language model for intent inference. Kardome says Cognition AI uses edge AI and distributed intelligence, processing voice recordings and biometric data locally. Directed speech detection is intended to distinguish interface speech from background conversation. VoiceID can be added to existing devices and applications, and the wider technology can complement ASR systems and platforms such as Amazon, Google, and Apple. It runs on Android and Linux.