Automating Archive Audio Workflows with Ortana

Ortana Media Group builds tailored solutions for data orchestration, provisioning, workflow and MAM through Cubix, its flagship platform. Cubix carries more than 170 API-driven integrations and a Media Aware Workflow Engine (MAWE), so customers assemble a workflow through configuration rather than in-house development.

Broadcasters and archive owners hold volumes of audio no team has the hours to process by hand — deep archive and LTO-based collections where dialogue sits buried in a finished mix, music rights are unclear, and metadata is thin. Ortana wanted to:

• Move content into audio processing without manual handoffs between systems
• Recover clean dialogue from finished mixes for intelligibility, localization and compliance
• Flag music usage and explicit lyrics before content reaches distribution
• Apply all of it across a whole library rather than one file at a time

Ortana connected AudioShake to Cubix Connect and its Media Aware Workflow Engine. Cubix ingests content from on-premise storage, cloud platforms, MAM/DAM systems and archive environments, then triggers AudioShake processing from workflow templates, rules or event-driven automation. Once a job completes, Cubix collects the separated audio and the metadata that comes with it and routes both back into the customer ecosystem, so results land wherever teams already work.

Broadcast archive compliance and localization. Cubix runs dialogue isolation to pull a clean speech track out of a finished mix, and multi-speaker separation to split overlapping voices for downstream editorial and processing. Nothing is generated: what comes back is recovered from the source recording. The same pass supports checks on whether audio remains in licence for reuse or distribution.

Lyric transcription and explicit tagging. Teams can run lyric transcription across music and mixed tracks, review the output, and flag explicit content. Cubix captures the result and writes tags back into the MAM/DAM, so a library carries consistent labels for editorial, brand safety and distribution compliance.

Enrichment at archive scale. Both workflows run en masse across large libraries, with the scheduling, monitoring and automation logic to handle high volume without overwhelming an operations team. That moves archive processing from an ad-hoc task to a repeatable one.

For Ortana customers, audio separation stops being a project and becomes a step in the pipeline — one that runs on the archive they already own, at the volume the archive actually holds.