#speech-recoginition

KanjuTech Transcription and Diarization

★ 1.3

KanjuTech Automatic Speech Recognition (ASR) model is a custom AI solution that generates meeting notes, subtitles, and other speech-to-text tasks. KanjuTech ASR model is distributed via the AWS Marketplace as a pre-built SageMaker Docker image.

Picovoice

★ 2.1

Leopard is an on-device speech-to-text engine offering private, accurate and affordable transcription experiences with zero latency. Leopard, with its small model size, can run anywhere from single-board computers such as Raspberry Pi, web browsers and servers.