> For the complete documentation index, see [llms.txt](https://docs.borndigital.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.borndigital.ai/insights/building-new-projects/analysis-project/3.-data-upload.md).

# 3. Data upload

Upload the recordings or texts to analyse, or connect a data source, and follow their processing.

Open **Upload** › **Data sources**. What you see depends on the project type. A project uses **one data source method**.

## Audio projects

1. Choose the speech-to-text **Provider** – **Azure**, **Azure Premium**, **ElevenLabs**, **Mistral** or **Gemini** – and the transcript **Display type**. Click **Save**.
2. Add the recordings:
   * **Upload audio files** – `.wav`, `.mp3`, `.ogg` or `.flac`, suitable for testing and smaller batches. Optional **Metadata** (JSON) is stored with the file and can be used in filters, conditions and dashboards.
   * **Copy data from another project** – reuse records of another Audio project, for example to test new parameters without touching the original.
   * **Sourcing from Azure** – recordings placed in the Azure storage of your environment are picked up automatically. Ask Born Digital to set it up.

<figure><img src="https://4261467870-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F6iQTvxgRZRwPS1NgIGEb%2Fuploads%2Fgit-blob-e1efa27fa715de0aac38e0b3711dbdc189411579%2Finsights-stt-provider.png?alt=media" alt="Data sources of an Audio project with the Provider list Azure, Azure Premium, ElevenLabs, Mistral and Gemini highlighted, Display type lexical and the Upload area for .wav, .mp3, .ogg and .flac files"><figcaption><p>Choosing the speech-to-text provider</p></figcaption></figure>

Details and the strengths of each provider: [Data sources (Audio)](/insights/workspace/upload/data-sources-audio.md).

## Text projects

* **Custom mailbox** – e-mails from your Office 365 mailbox (**Configure O365**).
* **Default mailbox** – e-mails sent or forwarded to the project's own address.
* **Upload** – a `.csv` file with one text per row; choose the column to analyse in the preview.
* **Send text** – type a single text (optionally with attachments and metadata), handy for testing.
* **Copy data from another project** – reuse records of another Text project.

Details: [Data sources (Text)](/insights/workspace/upload/data-sources-text.md).

## Follow the processing

New records are processed automatically: transcribed (Audio projects) and analysed with your prompts.

* **Process** › **Unprocessed** shows the records that are still waiting or in progress, with their status and any failures.
* **Process** › **Batch jobs** shows larger jobs such as reruns.
* The finished results appear in the dashboards – see the next step.

When you change parameters or prompts later, analyse existing records again with **Process** › [**Rerun**](/insights/workspace/process/rerun.md).