---
title: "Audio"
description: "Content people hear rather than read: podcasts, voice notes, interviews, narration, audio clips and spoken interfaces."
category: "Editorial craft"
canonical: "https://contentlexicon.com/audio"
updated: "2026-10-06"
---

# Audio

_Editorial craft_

> Content people hear rather than read: podcasts, voice notes, interviews, narration, audio clips and spoken interfaces.

Audio can be intimate, immediate and expressive in ways written content is not, because [voice](https://contentlexicon.com/voice) carries pace, emphasis, hesitation, warmth and authority.

For [content teams](https://contentlexicon.com/content-team), audio isn't “the same words, but with a microphone”. Spoken content has its own rhythm: sentences need room to breathe and the structure has to be audible. Listeners can't skim. Nor can they glance back up the page for a heading they missed, so signposting, pacing and repetition all work differently.

## Why it matters for content

Audio gives organisations another way to explain, teach, interview, reflect and build a relationship with an audience. It can be especially useful where human voice adds trust or texture: interviews, expert conversations, stories, oral histories, guided explanations, internal updates or material people may want to consume while doing something else.

It also brings responsibilities. Audio content needs planning, [editing](https://contentlexicon.com/editing), transcripts, summaries, ownership and a place in the wider [channel strategy](https://contentlexicon.com/channel-strategy). A 47-minute recording with no transcript, no useful notes and one point buried at minute 38 is an obstacle course with headphones.

## How AI changes the picture

Synthetic narration can offer another route through written source material, letting someone listen to a report or explanation they might otherwise struggle to make time for. AI can also help transcribe, summarise and edit recorded audio. The result still needs the rhythm and structure of spoken content, rather than a page read aloud without thought for the listener.

Listening can become responsive. An interactive audio experience can let someone interrupt to ask what an unfamiliar term means or request more detail before continuing. Google's December 2024 announcement of interactive Audio Overviews in NotebookLM demonstrated this direction: listeners could join an exchange with AI hosts to ask a question. It was an experimental feature, not evidence that every audio service can offer reliable conversation.

Recorded human testimony and generated speech are different kinds of source, and listeners should be able to distinguish them. A real person's voice shouldn't be cloned without their consent, and synthetic voices need disclosure. Editorial checks and text alternatives remain part of the experience: someone should be able to read the material, check a claim and understand what they're hearing.

## Questions people ask

**Is audio mainly about podcasts?**

Podcasts are one form. Audio also includes interviews, narration, audio articles, voice notes, recorded talks, guided experiences and voice interfaces.

**Does audio need a transcript?**

Yes, for anything you publish: [WCAG](https://contentlexicon.com/wcag) expects a text alternative for recorded audio. Transcripts also support search, reuse, quotation and people who prefer to read.

**Should every article have an audio version?**

Not necessarily. It depends whether listening adds value for the audience and whether the organisation can maintain the format properly.

**Can AI make audio content from written content?**

Technically yes, but that doesn't automatically make it good audio. Spoken content still needs rhythm, structure, editing and a reason to exist.

## Learn more

- [Making audio and video media accessible](https://www.w3.org/WAI/media/av/) – W3C guidance on transcripts, captions and accessible media.
- [Campfire tales: The essentials of writing for radio](https://www.npr.org/sections/npr-training/2025/05/31/g-s1-65875/campfire-tales-the-essentials-of-writing-for-radio) – NPR Training on short sentences, pacing and letting the recorded voices do the work.
- [NotebookLM gets a new look, audio interactivity and a premium version](https://blog.google/innovation-and-ai/models-and-research/google-labs/notebooklm-new-features-december-2024) – Google's December 2024 announcement shows listeners joining an Audio Overview to ask its AI hosts for explanation, with the feature's experimental limitations made explicit.

Related terms: [Accessibility](https://contentlexicon.com/accessibility), [Channel strategy](https://contentlexicon.com/channel-strategy)

This page was last reviewed in October 2026.