Direct-to-Transcript Overview
Direct-to-Transcript (DtT) is a lightweight call processing platform that facilitates real-time and post-call transcription. The DtT solution uses the following components:
V-Blaze — DtT uses V‑Blaze's Automatic Speech Recognition (ASR) engine to transcribe audio.
OrecX Audio — The OrecX Audio capture service provides call audio and metadata for Direct-to-Transcript. It supports active and passive VoIP recording.
Voci Adapter — Voci Adapter is the middleware linking OrecX Audio, V‑Blaze's ASR engine, and the endpoint to which utterances, transcripts, and audio are delivered.
OrecX Web (optional) — OrecX Web (orkweb) is a web application that stores call audio and associated metadata in a database. OrecX Web also provides a user interface to manage and view the files received from OrecX Audio.
![]() |
OrecX Audio establishes a connection to a telephony network to capture audio and metadata. Voci Adapter manages the following processes:
Provide the audio and metadata captured by OrecX Audio to V‑Blaze for transcription.
Deliver transcribed utterances and metadata to an endpoint in real time.
Deliver a complete transcript along with the associated audio file and metadata to the desired endpoint when a call is complete.
Direct-to-Transcript Deployment Outline
The sections in this guide detail the Direct-to-Transcript installation process as outlined below:
Important
The topics discussed in this guide assume you have a CentOS 7 system, V‑Blaze, and an OrecX license.