> For the complete documentation index, see [llms.txt](https://usermanual.doctranslate.io/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://usermanual.doctranslate.io/ai-office/meetings-and-notes.md).

# Meetings & Notes

**The AI ​​Meeting Assistant** feature automatically converts meeting content from audio files into written records. The system utilizes AI to identify speakers, transcribe speech to text, summarize content, and generate minutes based on either pre-existing or user-provided templates, thereby saving time on note-taking and post-meeting information consolidation.

## User Manual

### Step 1: Upload audio file

In the Audio File section, drag and drop or click to select the meeting audio file from your computer.\
Supported formats: MP3, WAV, M4A, FLAC, AAC, OGG, WebM.\
Maximum file size: 300 MB.

<figure><img src="/files/mrssgGT7duANYO8RGUaZ" alt=""><figcaption></figcaption></figure>

### Step 2:Declare speaker (Optional)

In the "Speakers" section, click "Add speaker" to enter details for each meeting participant.
\
For each speaker, you can enter:
\
\- Full name.
\
\- Gender.
\
\- Description or role (e.g., Department Manager, Client, Sales Representative...).
\
If this information is not provided, the AI ​​will automatically identify and distinguish the speakers during processing.

<figure><img src="/files/zgWaHYoCZj9I7HIUwJuO" alt=""><figcaption></figcaption></figure>

### Step 3: Provide context (Optional)

In the Context section, provide information that helps the AI ​​understand the meeting content more accurately.

Users can choose one of two methods:

* Enter text: Input information such as project names, programs, proper nouns, technical terms, or related content.
* Upload file: Upload reference documents containing information to support the meeting.

Contextual information helps the AI ​​correctly identify technical terms and improves the accuracy of the transcript.

<figure><img src="/files/AZfpr55qbnPnEEiV9QL3" alt=""><figcaption></figcaption></figure>

### Step 4: Select template (Optional)

In the "Template" section, drag and drop or select a template file to format the output record.
\
The system supports:

* PDF
* DOCX

If a template is not uploaded, the system will use the default minutes template.

### Step 5: Start a conversation

After completing the settings:

* Click "Create conversation" to have the system begin processing.
* AI will automatically convert audio to text, distinguish between speakers, and generate conversation content.

### Step 6: Review the conversation

Once processing is complete, proceed to the conversation check step.

Here, users can:

* Review the transcribed content.
* Check the speaker separation.
* Edit or add content as necessary before generating the transcript.

### Step 7: Generate and download the minutes

After confirming the content of the conversation:

* Proceed to the "Minutes" step.
* The system will automatically generate the content based on the selected or default template.
* Review the minutes and download them for storage or sharing.