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

> You choose the language the AI uses for each interview flow. On voice/video, you also choose the spoken voice (including accent / region).

# Multilingual Support

**PDF reports don’t have to stay in one language.** Reports follow the interview language you set in **Essential Settings** — and multilingual flows produce PDFs in that language. To download, select the candidate(s) from the list, click **Download**, and choose **Bulk PDF report (ZIP)** or **Package ZIP**. See [Reviewing Candidates](../../getting-started/log-in/reviewing-candidates#4-candidate-reports).

Need a language we don’t offer yet? Email [support@heymilo.ai](mailto:support@heymilo.ai).

## How it works

* **Interview language**: What the AI speaks and types in (questions, follow-ups, instructions). Set in **Essential Settings** as **Default Interviewer Language**.
* **Voice & accent**: Only for **Voice/Video**. You pick a specific voice after you add options from **LMNT** or **Cartesia**; accents line up with **language + region** (e.g. English → US, UK, India, Australia, Singapore, New Zealand — whatever appears in that language’s list).
* **Auto-detection (beta)**: Optional. Interview **starts** in your default language; if the candidate answers in **another** language, the AI can **switch** — behavior is spelled out in the product note next to the toggle.
* **Workspace vs one agent**: **Interviewers → Tools → Languages & Voices** is your catalog and defaults; **Voice Interview → Select a Voice** is where you lock in voice for that step.

## Where to click

| Task                                                       | Path                                                                                                             |
| ---------------------------------------------------------- | ---------------------------------------------------------------------------------------------------------------- |
| Browse languages, preview voices, see **Workspace Voices** | **Interviewers → Tools → Languages & Voices** — use search, expand a language, **play** to preview               |
| Add voices from a provider                                 | Same page → **+ Browse Providers** → **LMNT** or **Cartesia** → **+ Add**                                        |
| Set language for the **whole** flow                        | **Essential Settings** → **Default Interviewer Language**                                                        |
| Turn on language switching (beta)                          | **Essential Settings** → **Multilingual Auto-Detection**                                                         |
| Set **voice + accent** for voice/video                     | **Voice Interview** → **Settings** → **Identity** → **Voice** → **Select a Voice**                               |
| **Download a report in a chosen language**                 | Select candidate(s) on the list → **Download** → **Bulk PDF report (ZIP)** or **Package ZIP** → **Start export** |

<img src="https://mintcdn.com/heymilo/1Msj86qZdXlvphdL/images/image-42.png?fit=max&auto=format&n=1Msj86qZdXlvphdL&q=85&s=7d414381a4905610215eeb94a54a7f5f" alt="Image" title="Image" className="mx-auto" style={{ width:"55%" }} width="642" height="1408" data-path="images/image-42.png" />

## Reports and downloads

From the candidate list:

1. Select the candidate(s) you want.
2. Click **Download**.
3. Choose **Bulk PDF report (ZIP)** for PDFs only, or **Package ZIP** if you also need resume, audio, and video.
4. Click **Start export** — the file is ready when the export job completes.

PDF language follows the **Default Interviewer Language** (and **Multilingual Auto-Detection** settings) for that interview flow. Set those in **Essential Settings** before candidates interview if you need a specific language.

To share a read-only preview instead of downloading, open the candidate profile and use **Share**. See [Reviewing Candidates](../../getting-started/log-in/reviewing-candidates#sharing-a-read-only-report).

## Voice & accent (voice/video)

Voices are supplied by **LMNT** and **Cartesia**. You’re not stuck with one provider — add what you need, then choose per agent.

* **More voices**: **Browse Providers**, preview, **Add** to the workspace.
* **Wrong sound for the role**: Change it on the **Voice Interview** step (**Select a Voice**), not only in the global Languages & Voices page.
* **Deeper defaults** (pacing, tone): [Languages & Voices](../../tools/ai-settings/languages-and-voices).

<Frame caption="Browse LMNT or Cartesia, then Add.">
  <img src="https://mintcdn.com/heymilo/_i4f7QvAeVXsbdDh/images/multilingual-browse-voice-providers.png?fit=max&auto=format&n=_i4f7QvAeVXsbdDh&q=85&s=0f3199d377713e00bb38062afbb10d43" alt="Browse Voice Providers" className="mx-auto" style={{ width:"63%" }} width="605" height="1024" data-path="images/multilingual-browse-voice-providers.png" />
</Frame>

<Frame caption="Select a Voice on the voice/video step.">
  <img src="https://mintcdn.com/heymilo/_i4f7QvAeVXsbdDh/images/multilingual-select-voice-drawer.png?fit=max&auto=format&n=_i4f7QvAeVXsbdDh&q=85&s=e0cdf4e569442365f296af3905432a60" alt="Select a Voice drawer" width="1024" height="747" data-path="images/multilingual-select-voice-drawer.png" />
</Frame>

### Walkthrough video

Same flow as [Voice/Video Interviews](voice-interviews).

<iframe className="w-full aspect-video rounded-xl" src="https://storage.googleapis.com/heymilo-pub-docs/videos/Multilingual-Walkthrough.mp4" allowFullScreen />

## Essential Settings (languages)

<Frame caption="Default language and Multilingual Auto-Detection (BETA).">
  <img src="https://mintcdn.com/heymilo/_i4f7QvAeVXsbdDh/images/multilingual-essential-settings-language.png?fit=max&auto=format&n=_i4f7QvAeVXsbdDh&q=85&s=7085ffc7dd2f1ef0dbf69535e4b6f552" alt="Default Interviewer Language and Multilingual Auto-Detection" width="1024" height="416" data-path="images/multilingual-essential-settings-language.png" />
</Frame>

| Field                            | What it does                                                                                 |
| -------------------------------- | -------------------------------------------------------------------------------------------- |
| **Default Interviewer Language** | Base language for this interview flow                                                        |
| **Multilingual Auto-Detection**  | Starts in that language; may follow the candidate if they switch languages (see in-app copy) |

## By interview type

| Type              | Note                                                                                                        |
| ----------------- | ----------------------------------------------------------------------------------------------------------- |
| **Voice/Video**   | Language + voice/accent; main place to tune how the AI sounds. → [Voice/Video Interviews](voice-interviews) |
| **SMS**           | Follows **Essential Settings** for language with the current SMS experience                                 |
| **Form / Resume** | Set **Default Interviewer Language** so generated questions and text match the audience                     |

## Quick fixes

* **Candidate heard the wrong language**: Check **Default Interviewer Language** and **Multilingual Auto-Detection**; confirm you saved the flow.
* **Accent doesn’t fit**: Re-open **Select a Voice** and pick another region under the right language; preview first.
* **Can’t find a voice**: **+ Browse Providers** and add it; it should appear for workspace use.
* **PDF isn’t in the language you want**: Check **Default Interviewer Language** and **Multilingual Auto-Detection** on the interviewer’s **Essential Settings** — report language follows how the interview was configured.

## Related

* [Languages & Voices](../../tools/ai-settings/languages-and-voices)
* [Creating an Interviewer](../../getting-started/log-in/creating-an-interviewer)
* [Configuring Your AI Interviewer](../../getting-started/log-in/configuring-your-ai-interviewer)