RunBeat Compliance User Guide

Docs → RunBeat Compliance

Overview

RunBeat Compliance is a compliance framework built around Ofcom's rules and the Ofcom Broadcasting Code. There is no single "compliance policy" — rather, the Broadcasting Code sets out a series of rules and expectations that UK broadcasters must follow. RunBeat Compliance helps stations meet those expectations by automatically transcribing audio tracks, scanning lyrics against forbidden words lists, classifying content by severity, and maintaining audit trails that demonstrate due diligence.

Compliance is available as a standalone service at radiocompliance.co.uk and is also integrated directly into RunBeat Pulse. You can use it independently or as part of the Pulse playout workflow.

Getting Started

Standalone access

Visit radiocompliance.co.uk and log in using your station's credentials. The compliance website authenticates against your station's user pool, so you use the same login as your playout system.

Via Pulse

If you use RunBeat Pulse, compliance checking is built in. Every music track uploaded to Pulse is automatically submitted for compliance analysis — no separate login or action required.

Guided Wizards — "Show Me How"

Every Compliance page includes a "🎓 Show me how" button in the sidebar. Click it to open a library of step-by-step guided walkthroughs that walk you through common tasks without leaving the app.

Each wizard guides you through the process one step at a time, with tips and explanations along the way. You can dismiss a wizard at any point by pressing Escape or clicking the ✕ button.

Available wizards

Wizard	What it covers
🎧 Upload Audio for Checking	Submit audio files for automated compliance analysis with AI-powered scanning.
🚫 Manage Forbidden Words	Customise your word filter list with custom words, phrases, and severity levels.
📋 Review Compliance Reports	Check your compliance history, view flagged items, and download PDF reports.
💳 Manage Your Subscription	View your current plan, usage, upgrade options, and billing details.

Wizards are a great way to learn the platform — especially for new users or when exploring features you haven't used before.

How Compliance Checking Works

When a track is submitted for compliance checking, the following process runs automatically:

1. Transcription — The track's vocal content is transcribed to text using speech-to-text technology.
2. Text normalisation — The transcript is normalised (lowercased, symbols replaced, obfuscations decoded) to ensure consistent matching.
3. Baseline filter — The normalised text is checked against the built-in baseline filter (see below).
4. Custom forbidden words — The text is then checked against your station's custom forbidden words list.
5. Scoring — Matched words are scored based on severity, repetition, and density.
6. Decision — A compliance outcome is produced: Allow, Review, Block, or Escalate.
7. Audit record — A full audit record is stored for every scan.

Baseline Ingestion Filter

The baseline filter is a built-in set of rules informed by the expectations of the Ofcom Broadcasting Code. It provides a minimum standard of content screening that applies to all stations. The baseline filter:

• Is not editable from the application — it is a code-level constant.
• Is not stored in the database — it cannot be modified or weakened by users.
• Applies identically across all stations and deployments.
• Runs before your custom forbidden words check.

Detection layers

The baseline filter uses five detection layers, applied in order:

1. Exact phrase match — Multi-word offensive expressions matched as complete phrases.
2. Exact word match — Individual words matched by category (profanity, sexual content, slurs, threats).
3. Obfuscation match — Common deliberate misspellings and character substitutions (e.g. "f*ck", "sh1t") mapped to their canonical forms.
4. Fuzzy match — Near-matches with conservative edit distances to catch minor spelling variations without excessive false positives.
5. Repetition & density — Repeated use of offensive words and high density of risk content increase the severity score.

Normalisation

Before matching, the transcript is normalised to catch obfuscation attempts:

• All text converted to lowercase.
• Symbol substitutions decoded (@ → a, $ → s, ! → i, 0 → o, etc.).
• Asterisks removed.
• Repeated characters collapsed (e.g. "fuuuck" → "fuck").
• Whitespace normalised.

Forbidden Words Management

In addition to the baseline filter, each station maintains its own custom forbidden words list. You can manage your list from the compliance website or from within Pulse:

• Upload — Upload a text file with one word or phrase per line. Optionally specify severity with a pipe delimiter (e.g. "word|major").
• Add/edit/remove — Manage individual words through the interface.
• Severity assignment — Each word is classified as "minor" or "major" (defaults to minor if not specified).

Your custom words layer on top of the baseline filter — you can add additional words but cannot remove or weaken baseline rules. Changes take effect immediately for all future compliance checks.

Severity Classification

Severity	Indicator	Automated action
Minor	● Amber	Track is held for admin review. Stored but unavailable for broadcast until reviewed.
Major	● Red	Track is rejected and automatically removed from the system.

When a track matches only minor-severity words, it is held for human review. When any major-severity word is matched, the track is automatically rejected and deleted.

Compliance Outcomes

Outcome	Status	What happens
Allow	Approved	No forbidden words found. Track is immediately available for broadcast.
Review	Held	Minor-severity words found. Track is stored but blocked from broadcast until an admin reviews it.
Block	Rejected	Major-severity words found. Track is automatically deleted.
Escalate	Escalated	Critical content (discriminatory slurs, threats, or very high risk scores). Flagged for immediate senior review with a distinct purple badge.

Review Queue

Tracks with a status of "held" appear in the compliance review queue. For each track, the review interface shows:

• Track title and artist.
• The full lyrics transcript with matched forbidden words highlighted.
• An audio preview player so you can listen to the track.
• The severity level and category of each matched word.

As an admin, you can:

• Approve — Mark the track as safe. It becomes available for broadcast.
• Reject — Mark the track as unsafe. It is deleted from the system.

All review decisions are logged in the audit trail with the reviewer's identity and timestamp.

Bulk actions

Select multiple held tracks using checkboxes (or "Select All") and approve or reject them in a single action — useful when processing a large batch upload.

Independent Track Checking

The compliance website allows you to upload tracks for standalone checking without importing them into your playout system. This is useful for:

• Pre-screening tracks before adding them to your library.
• Checking content from external sources.
• Verifying compliance without affecting your playout workflow.

Upload audio files (MP3, WAV, or FLAC, up to 50 MB) and receive the full compliance result including status, matched words, severity levels, and the complete transcript. Independent check results are clearly labelled as "Independent Check — not linked to playout import".

Results are retained for 90 days (audio file deleted after 90 days, but the compliance decision record is kept for 2 years).

Compliance Reports

Generate compliance decision reports for regulatory review. Reports can be filtered by date range or by the most recent upload batch, and include:

• Track title and artist.
• Upload date.
• Compliance decision (approved, held, rejected, escalated).
• Matched forbidden words with severity levels.
• Reviewer name and review date (for manually reviewed tracks).

Reports can be downloaded as PDF or emailed directly to a specified address. Each report includes the station name, generation date, date range, and the compliance disclaimer on every page.

Audit Trail

Every compliance scan produces a structured audit record containing:

• Track identifier, title, and artist.
• Decision (Allow/Review/Block/Escalate).
• Total risk score.
• All matched rules with occurrence counts and score contributions.
• Processing timestamp.
• Input source (upload, independent check, etc.).

Audit records are retained for a minimum of 2 years and cannot be deleted during this period. This ensures your station can demonstrate due diligence under the Ofcom Broadcasting Code at any time.

Pulse Integration

When used with RunBeat Pulse, compliance checking is fully automatic:

• Every music track uploaded to Pulse is automatically submitted for compliance analysis.
• Compliance status badges appear on all tracks in the media library.
• Tracks that are processing, held, or escalated are automatically blocked from scheduling and the play log.
• The compliance review queue is accessible directly within the Pulse interface.
• Non-music categories (jingles, adverts, sweepers) bypass compliance checking and are marked as "not applicable".

Compliance Disclaimer

This disclaimer appears on the compliance website landing page, every compliance report, every independent upload result, and the compliance review screen within Pulse.

Subscription Tiers

Tier	Monthly	Checks/month	Features
Free	£0	3	Basic compliance checking, on-screen results.
Professional	£29	100	Full transcription, severity classification, PDF reports, email delivery.
Broadcast	£79	500	Everything in Professional plus priority processing, bulk operations, and 2-year audit retention.

Additional compliance checks can be purchased as add-ons (100-pack for £9/month) from the Professional tier upward.