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

# AI-Generated Audio Detection

> Classify audio as AI-Generated or human-created.

Deploy's AI-Generated Audio Detection API analyzes audio and video content to determine whether the audio is AI-generated. The API returns classification confidence scores for each analyzed segment, helping identify potentially artificial audio content.

The API processes media in 10-second audio chunks and returns timestamped classification results, allowing you to identify AI-generated audio within longer recordings.

## Detection Capabilities

> [card] AI-Generated Audio Detection
>
> Determine whether an audio recording is AI-generated or human-created, with confidence scores for each classification.

## AI-Generated Audio Detection

The AI-Generated Audio Detection model determines whether an audio recording was generated using AI. It is designed to distinguish AI-generated audio from human-created audio.

The model contains one classification head with two possible classes.

### Classification Outputs

| Classification | Description |
| - | - |
| `ai_generated` | Confidence that the audio is AI-generated |
| `not_ai_generated` | Confidence that the audio is not AI-generated |

Each classification returns a confidence score between 0 and 1, indicating the model's confidence in its prediction. The confidence scores within this classification head sum to 1.

## Detection Results and Response Structure

The API divides each input audio or video file into 10-second audio chunks and returns a classification for each segment.

Each classification includes:

* **Timestamp:** The start time of the analyzed audio segment.
* **Classification:** Whether the segment is AI-generated or not AI-generated.
* **Confidence scores:** The model's confidence for each classification.

If the submitted audio is shorter than 10 seconds, the API returns a single classification for the entire clip.

## Recommended Detection Thresholds

We recommend the following thresholds as starting points for production workflows.

| Detection Objective | Recommended Threshold |
| - | - |
| General Accuracy | 0.90 |
| Recall | 0.50 |
| Precision | 0.99 |

> [!NOTE]
>
> These thresholds are general recommendations. For optimal performance, calibrate them using representative data from your use case and your tolerance for false positives and false negatives.

## Supported File Formats

> [tab] Video
>
> * MP4
> * WEBM
> * AVI
> * MKV
> * WMV
> * MOV

> [tab] Audio
>
> * FLAC
> * MP3
> * OGG
> * WAV
> * M4A

## API Reference

> [!NOTE]
>
> This API is currently only available on our legacy API. Please refer to our [AI-Generated Audio Detection API Reference](https://docs.thehive.ai/reference/ai-generated-audio-detection-1) for more details.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.