> ## Documentation Index
> Fetch the complete documentation index at: https://docs.unleashx.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Transcription

> Turn recorded calls and audio files into accurate text transcripts.

# Transcription

> **In one line:** Convert any recorded audio (a call, a voice note, an uploaded file) into a clean text transcript your agent can use.

|                    |                                                                        |
| ------------------ | ---------------------------------------------------------------------- |
| **Category**       | AI & Media                                                             |
| **Authentication** | Platform-managed                                                       |
| **Setup time**     | \~1 minute                                                             |
| **Difficulty**     | Easy                                                                   |
| **Best for**       | Transcribing calls and audio for summaries, search, QA, and follow-ups |

## 1. Overview

Transcription (speech-to-text) converts spoken audio into written text. Give it the URL of an audio file and it returns the words that were said, smart-formatted and ready to use.

Once connected, your agent can transcribe recorded calls, voicemails, and uploaded audio, choose between speech providers (Deepgram or Azure), and target a specific language when you know it. The result is plain text you can summarize, search, score, or store.

Connecting Transcription to UnleashX lets your voice and automation agents turn raw audio into structured insight — call summaries, sentiment, action items — without anyone listening back manually.

## 2. What you'll need

<Note>
  Transcription is **built into UnleashX**. There are no third-party accounts (Deepgram/Azure) for you to create and no API keys for you to manage.
</Note>

* An active UnleashX account.
* The Transcription feature enabled on your workspace/plan.
* A publicly reachable URL to the audio file you want transcribed.
* Permission to edit the agent (admin or editor role). Without it, ask a workspace admin to enable the feature.

## 3. Get your credentials

<Info>
  There are **no credentials to create**. Transcription is platform-managed — UnleashX provisions and rotates the underlying Deepgram and Azure Speech keys. You never handle a provider key.
</Info>

If you are a workspace admin, the only optional platform-side choices are which provider and default model/language to use. UnleashX ships sensible defaults; most teams never change them.

| Platform-managed setting    | Plain-English reason it exists                                       |
| --------------------------- | -------------------------------------------------------------------- |
| Provider (Deepgram / Azure) | Lets UnleashX route audio to the speech engine that fits your needs. |
| Default model & language    | Sets transcription quality and the expected spoken language.         |
| Provider API keys           | Provisioned and rotated by UnleashX so transcription "just works."   |

## 4. Connect on UnleashX

<Steps>
  <Step title="Open your agent">
    Go to [https://www.tryunleashx.com](https://www.tryunleashx.com) and open the agent that should transcribe audio.
  </Step>

  <Step title="Open Data Connectors">
    In the agent, click **Data Connectors**.
  </Step>

  <Step title="Find Transcription and add it">
    Locate **Transcription** and click **Connect** / **Add**. It's platform-managed, so there's no key to paste — it activates immediately.
  </Step>

  <Step title="Confirm it's connected">
    The Transcription tool should now show a **Connected** badge.
  </Step>
</Steps>

### Use Transcription in a Workflow

Once connected, you can add **Transcription** to any automation from the **Workflows** builder. Its triggers and tools appear in the Apps panel, marked with an **MCP** badge.

<Steps>
  <Step title="Add a trigger or action node">
    Open **Workflows → New Workflow**. On the canvas, click **+ Add Trigger** (or the **+** below any node) to add a step.

    <img src="https://mintcdn.com/unleashx/LXZeho1M8XzFkz0k/images/mcp/workflows/05-add-trigger.png?fit=max&auto=format&n=LXZeho1M8XzFkz0k&q=85&s=449a0c03b0c7b40e571ec02373a85259" alt="Add a trigger on the workflow canvas" width="1390" height="722" data-path="images/mcp/workflows/05-add-trigger.png" />
  </Step>

  <Step title="Pick Transcription from the Apps panel">
    In the **Paths** panel, open the **Apps** tab and select **Transcription**. Use the search box if you have many connectors.

    <img src="https://mintcdn.com/unleashx/LXZeho1M8XzFkz0k/images/mcp/workflows/06-apps-list.png?fit=max&auto=format&n=LXZeho1M8XzFkz0k&q=85&s=bbb8bb6e668c4f8e1f6fd5e4955a45f0" alt="Choose your app from the Apps list" width="2446" height="1550" data-path="images/mcp/workflows/06-apps-list.png" />
  </Step>

  <Step title="Choose a trigger or tool">
    Pick the trigger or action you want. Configure its fields — required fields are marked with a red asterisk (\*).
  </Step>

  <Step title="Add or select your account">
    Under **Selected account**, choose an already-connected account, or click **Add Account** to connect one now.
  </Step>

  <Step title="Save and test">
    Fill in the remaining fields and click **Save**. Use **Test** to verify the step, then toggle **Publish** when the workflow is ready.
  </Step>
</Steps>

<Note>
  The steps are the same for every connector. For the full workflow builder guide, see [Using MCP in Workflows](/mcp/workflows).
</Note>

## 5. Available tools

| Tool             | What it does                                                     | Changes data? |
| ---------------- | ---------------------------------------------------------------- | ------------- |
| Transcribe Audio | Transcribe audio from a URL using the Deepgram or Azure provider | No            |

<Note>
  Transcription is read-only — it reads an audio file and returns text. It does not alter or store the source recording.
</Note>

## 6. Example usage

**"Transcribe this call recording and give me a summary."**
→ Runs **Transcribe Audio** on the recording URL, then the agent summarizes the returned transcript.

**"Get the text of this Hindi voicemail."**
→ Runs **Transcribe Audio** with the language set to Hindi (`hi-IN`) so the transcript comes back accurately.

## 7. Permissions & data access

**UnleashX can:**

* Download the audio file at the URL you provide.
* Send it to the configured speech provider and return the transcript.
* Temporarily process and convert the audio (e.g. to a compatible format).

**UnleashX cannot:**

* Access audio you don't explicitly pass to the tool.
* Permanently store your recordings as part of the integration.
* Record new audio on its own.

**To disconnect:** Open the agent → **Data Connectors** → **Transcription** → **Disconnect**. Access is revoked immediately.

## 8. Troubleshooting

| Problem                                 | What it means                                       | How to fix it                                                       |
| --------------------------------------- | --------------------------------------------------- | ------------------------------------------------------------------- |
| "url must be a valid http or https URL" | The audio link is missing or malformed              | Pass a full, public `http(s)` URL to the audio file                 |
| "Audio file too large"                  | The file exceeds the 50 MB limit                    | Trim or compress the recording, or split it                         |
| 401 / credential error                  | The platform-managed provider key is unavailable    | Platform-side — contact **[cs@unleashx.ai](mailto:cs@unleashx.ai)** |
| 403 / feature not enabled               | Transcription isn't enabled on your plan            | Ask a workspace admin or contact support                            |
| Empty transcript                        | Silent audio, wrong language, or unsupported format | Confirm the file has speech and set the right language              |

For general connector issues, see [/mcp/integrations](/mcp/integrations).

## 9. Frequently asked questions

**Is my audio stored?**
No. The file is downloaded only to produce the transcript and is not retained as part of the integration. Temporary working files are cleaned up.

**Do I need a Deepgram or Azure account?**
No. Transcription is platform-managed — UnleashX provides and rotates the provider keys.

**Which provider is used?**
Deepgram or Azure, depending on your workspace configuration. You can pass `stt_provider_name` to choose per request.

**Can multiple team members use it?**
Yes. Once enabled on the workspace, anyone with access to the agent can transcribe audio.

## 10. References

* Deepgram speech-to-text docs: [https://developers.deepgram.com/docs](https://developers.deepgram.com/docs)
* Azure Speech service docs: [https://learn.microsoft.com/azure/ai-services/speech-service/](https://learn.microsoft.com/azure/ai-services/speech-service/)
* UnleashX integrations help: [/mcp/integrations](/mcp/integrations)