Zoom ships with several captioning options out of the box. For many meetings, they do the job. But if you've ever tried to push French accents, the Spanish ñ, or a cédille through Zoom's built-in caption software, you'll know the experience quickly falls apart. Characters either fail to render or appear as garbled symbols, and that's a problem that no amount of refreshing will fix.

For AV teams managing multilingual corporate events, webinars, or international conferences on Zoom, this limitation isn't a minor inconvenience. The solution is Zoom's API token live captioning method, which bypasses the built-in interface entirely and allows a third-party captioning platform - like Line 21 - to stream properly formatted captions directly into the meeting.

This guide walks through exactly how that works: why the built-in caption offering falls short, what the API token method enables, and how to set up Zoom as your Line 21 captioning destination step by step.

Zoom's Built-In Captioning Options: A Quick Overview

Before diving into the API token setup, it helps to understand what Zoom offers natively. There are three main captioning paths available within the platform:

1. Automated live transcription: Zoom's AI transcription, enabled at the account level, which generates captions automatically. Generally poor quality for accuracy.
2. Manual captioner: A feature that allows the host to assign a participant to type captions live directly into Zoom's interface.
3. Third-party captioning via API token: A REST API-based method that allows external captioning software to stream text directly into the Zoom session, whether human or automated.

For basic English-language meetings, options one and two are often sufficient. The problems start when your requirements go beyond that.

The Limitations of Zoom's Built-In Captions

Zoom's default automatic speech recognition (ASR) captions, while convenient, often fall short of the accuracy and versatility required for professional environments. These captions rely on automated algorithms to transcribe speech into text in real time, but the output is frequently riddled with errors, particularly in scenarios that involve multiple speakers, technical jargon, or diverse accents.

For professionals in the Audio-Visual (AV) space, where precision is critical, these inaccuracies can result in miscommunication and a poor experience for participants. Furthermore, Zoom's ASR captions struggle to meet accessibility standards for events involving specialised terminology or multilingual participants, as the system doesn't offer comprehensive language support or robust customisation options.

While suitable for basic, informal meetings, Zoom's built-in captions lack the reliability and customisation capabilities that AV teams require for seamless corporate events, live broadcasts, or high-stakes productions. For situations demanding professional-grade performance, these shortcomings highlight the need for more advanced third-party solutions.

Zoom's manual captioning interface was designed for simplicity. A participant is assigned the captioning role, types into a small text box, and those captions appear at the bottom of participants' screens. It works - to a point.

The core issue is character support. Zoom's built-in captioner uses a constrained input format that doesn't reliably handle extended character sets. Accented characters common in French (é, è, ê, ç), Spanish (ñ, á, ü), or Portuguese (ã, õ) can fail to pass through correctly. The French cédille — the ç in words like façade or leçon — is a common culprit. It either doesn't render, displays incorrectly, or gets dropped entirely.

For a professional captioner, this creates a significant practical problem. They may be typing accurate, properly formatted captions on their end, only for participants to see broken text or missing characters. The caption quality that was promised simply doesn't arrive.

There's also the matter of caption format control. The built-in interface offers limited flexibility in how captions are structured, timed, or displayed — all of which matter for professional-grade captioning services.

Why the API Token Method Solves This

Zoom's API token captioning method works differently. Instead of relying on the in-meeting text interface, a third-party captioning platform connects to a unique URL, the API token - and sends live caption data directly to Zoom as UTF-8 encoded plain text via HTTP POST requests.

Because the data travels as UTF-8, the full range of Unicode characters is supported. That means French cédilles, Arabic script, Cyrillic characters, and other extended character sets all pass through correctly, provided the captioning platform handles encoding properly. Line 21 does.

Beyond character support, the API token approach gives you genuine control over the captioning pipeline. You can use a professional human captioner or an ASR engine of your choice. You can route captions through Line 21's AI Proofreader for real-time accuracy improvements. And you retain full flexibility over how and where captions are delivered — including multiple destinations simultaneously if needed.

One important constraint to be aware of: Zoom's API token only supports a single language at a time per session. If you need multilingual captions across different audiences, you'll need to plan your session architecture accordingly.

Part 1: Setting Up API Token Captioning in Zoom

Before Line 21 can connect to your meeting, you need to enable third-party captioning in your Zoom account settings. This must be done via the Zoom web portal - the Zoom desktop app does not expose these settings.

Step 1: Enable Manual Captions in Zoom Settings

1. Sign in to your Zoom account at zoom.us using a browser.
2. Navigate to Settings → Meeting → In Meeting (Advanced).
3. Scroll to Manual Captions and toggle it on.
4. Ensure both of the following options are enabled:
   • Allow host to type or assign a participant to type
   • Allow use of caption API Token to integrate with 3rd-party Closed Captioning services
5. Optionally, enable Full Transcript and Save Captions if you want participants to access a running transcript or save a copy after the session.

Once this is configured at the account level, it applies to all meetings you host. You won't need to repeat this step each time.

Step 2: Retrieve the API Token During Your Meeting

The API token is generated live, it only becomes available once your meeting has started, and it expires when the session ends. If the meeting restarts for any reason, a new token is issued.

To retrieve it:

1. Start your Zoom Meeting or Webinar as the host.
2. In the meeting toolbar, click the up arrow next to the CC / Show Captions button.
3. Select Set up manual captioner.
4. Click Copy the API token.

The token takes the form of a URL — something like:

https://wmcc.zoom.us/closedcaption?id=200610693&ns=GZHkEA==&expire=86400&...

This is the URL you'll paste into Line 21 to establish the connection. Share it securely with your captioning operator, it can also be sent via Zoom's in-meeting chat if needed.

Important: Do not enable Zoom's automated live transcription at the same time as your third-party captioning connection. The two will compete, producing illegible, overlapping captions for your participants.

Part 2: Connecting Line 21 to Zoom via API Token

With the API token in hand, the next step is configuring Line 21 to use Zoom as its caption destination.

What Line 21 Needs Before the Session

Before the session starts, your captioning setup should be confirmed:

• If using a human captioner: The captioner needs to join the Zoom call directly so they can hear the audio in real time. They'll caption from within Line 21 while connected to the meeting as a participant.
• If using automated ASR captioning: Line 21's AI Call Agent joins the Zoom session to receive the audio stream and generate captions automatically.

In either case, the captions are then routed from Line 21 to Zoom via the API token, not typed directly into Zoom's interface.

Setting Up the Zoom Destination in Line 21

In your Line 21 session, navigate to the Destinations section and select Zoom as your text destination. From there:

1. Name the session — give it a descriptive label so it's easy to identify in your Line 21 dashboard.
2. Enter the API token URL — paste the URL copied from Zoom into the designated field.
3. Select the caption language — choose the language your captions will be delivered in.

Once saved, Line 21 is ready to stream captions into Zoom as soon as captioning begins.

Part 3: Running the Full Captioning Workflow

Here's how the end-to-end process looks on the day of your event.

Before the session:

• Confirm your Zoom account settings have API token captioning enabled.
• Ensure your Line 21 session is created, with the captioner or ASR engine assigned.
• Brief your Zoom meeting host on the token retrieval steps.

At session start:

1. The Zoom host starts the meeting and retrieves the API token (as described above).
2. The token is shared with the Line 21 operator or captioner.
3. The operator enters the token into the Zoom destination in Line 21.
4. The human captioner joins the Zoom session as a participant, or the AI Call Agent is enabled for ASR and invited into the meeting.

During the session:

• Captions begin streaming from Line 21 into Zoom in real time.
• Participants can toggle captions on or off using the CC button in their Zoom toolbar.
• If you've enabled Full Transcript, participants can also view a running text panel alongside the meeting.
• The Zoom host does not need to interact with the captioning system once it's live.

A few things to keep in mind:

• Breakout rooms each require their own separate API token — plan ahead if your session uses them.
• If the meeting drops and restarts, a new token must be retrieved and re-entered in Line 21.
• Extended character sets — including the French cédille and other diacritics — will render correctly for participants, provided Zoom's automated captions are disabled.

Troubleshooting Common Issues

Even with a clean setup, things can occasionally go wrong. Here are the most common issues and how to resolve them:

Getting Started With Line 21 for Zoom Captioning

The API token method is the right approach any time your captioning requirements go beyond basic English-language transcription. Whether you're running a bilingual corporate town hall, a multilingual conference, or any session where character accuracy matters, the built-in Zoom interface simply won't hold up under professional scrutiny.

Line 21 is built for exactly these workflows. It supports human captioners and ASR engines, delivers captions with full Unicode character support, and integrates directly with Zoom's REST API for a stable, low-friction connection. For AV teams running regular events, it's a setup that scales from a single session to a full event calendar without adding operational complexity.

To configure your first Zoom session with Line 21, visit know.line-21.com/docs or get in touch with the Line 21 team to discuss your specific requirements.