---
license: cc-by-4.0
language:
- en
pretty_name: TartanAviation ATC + ADS-B
task_categories:
- automatic-speech-recognition
tags:
- atc
- ads-b
- aviation
- air-traffic-control
- speech
size_categories:
- 10K<n<100K
configs:
- config_name: default
  data_files:
  - split: train
    path: shard-*.parquet
dataset_info:
  features:
  - name: clip_id
    dtype: string
  - name: airport
    dtype: string
  - name: date
    dtype: string
  - name: start
    dtype: timestamp[us]
  - name: end
    dtype: timestamp[us]
  - name: duration_s
    dtype: float64
  - name: n_aircraft
    dtype: int32
  - name: tails
    list: string
  - name: tracks
    list:
    - name: t
      dtype: timestamp[us]
    - name: tail
      dtype: string
    - name: aircraft_id
      dtype: string
    - name: lat
      dtype: float64
    - name: lon
      dtype: float64
    - name: alt
      dtype: int32
    - name: speed
      dtype: int32
    - name: heading
      dtype: int32
  - name: audio
    dtype:
      audio:
        sampling_rate: 16000
---

# TartanAviation ATC + ADS-B

Paired ATC audio and ADS-B for Pittsburgh KAGC and KBTP, aligned from CMU
[TartanAviation](https://theairlab.org/tartanaviation/). Each row is one ADS-B-triggered
audio capture (16 kHz mono) plus the aircraft tracks present during it.

41,823 clips · 16 kHz mono · 67% carry ADS-B. Built with [squawk](https://github.com/twangodev/squawk).

## Usage

```python
from datasets import load_dataset

ds = load_dataset("twangodev/tartanaviation-atc-adsb", split="train", streaming=True)
ex = next(iter(ds))
ex["audio"]    # {'array': ..., 'sampling_rate': 16000}
ex["tails"]    # callsigns present, e.g. ['EJA660', 'RPA4996']
ex["tracks"]   # per-ping lat/lon/alt/speed/heading/tail in the clip window
```

## Schema

| column | type | notes |
|---|---|---|
| `clip_id` | string | `{airport}/{date}/{n}` |
| `airport` | string | `kagc` / `kbtp` |
| `date` | string | `MM-DD-YY` |
| `start` / `end` | timestamp | clip window (audio and ADS-B share one clock) |
| `duration_s` | float | clip length |
| `n_aircraft` / `tails` | int / list | aircraft in-window (`0` / `[]` when no ADS-B) |
| `tracks` | list of struct | every ADS-B ping in-window: `t, lat, lon, alt, speed, heading, tail` |
| `audio` | `Audio(16000)` | 16 kHz mono WAV |

Clips are 2–14 min ADS-B-triggered captures, not VAD utterances (~85% is silence).

## License

CC-BY-4.0, inherited from CMU AirLab's TartanAviation. Credit CMU AirLab and cite:

```bibtex
@article{patrikar2024tartanaviation,
  title={TartanAviation: Image, Speech, and ADS-B Trajectory Datasets for Terminal Airspace Operations},
  author={Jay Patrikar and Joao Dantas and Brady Moon and Milad Hamidi and Sourish Ghosh and Nikhil Keetha and Ian Higgins and Atharva Chandak and Takashi Yoneyama and Sebastian Scherer},
  year={2024},
  eprint={2403.03372},
  archivePrefix={arXiv},
  primaryClass={cs.LG},
  url={https://arxiv.org/pdf/2403.03372.pdf}
}
```