File size: 1,966 Bytes
07af0d9
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1b1e8bb
07af0d9
1b1e8bb
07af0d9
 
 
1b1e8bb
07af0d9
 
1b1e8bb
 
 
 
 
 
 
 
 
 
 
 
 
07af0d9
 
 
1b1e8bb
07af0d9
1b1e8bb
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
07af0d9
 
 
 
 
 
 
 
 
 
 
 
 
 
1b1e8bb
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
---
license: unknown
task_categories:
- text-to-audio
- audio-to-audio
language:
- en
tags:
- audio
- spatial-audio
- first-order-ambisonics
- foa
- pcm16
- parquet
---

# PhysWave Dataset

Pre-rendered first-order ambisonic (FOA) release derived from Mono-Label-120k.
Each mono source clip is rendered twice: one static trajectory and one
deterministic moving trajectory (`linear`, `circular`, or `approach_recede`).

Audio is stored in parquet rows as an `audio` struct with embedded PCM16 WAV
bytes and a relative path. This audio column contains 4-channel FOA.

FOA convention:

```text
channel order: W, X, Y, Z
W = s / sqrt(2)
X = s * cos(elevation) * cos(azimuth)
Y = s * cos(elevation) * sin(azimuth)
Z = s * sin(elevation)
distance gain = 1 / max(distance_m, 0.5)
moving trajectories use retarded-time propagation at 343 m/s
```

Trajectory waypoints are stored as JSON strings containing 10 Hz arrays:
`[t_seconds, azimuth_deg, elevation_deg, distance_m]`. A 10 second clip has
101 waypoints.

Packing summary:

```json
{
  "source_records": 127269,
  "rendered_rows": 254538,
  "static_rows": 127269,
  "moving_rows": 127269,
  "missing_source_records": 0,
  "moving_type_counts": {
    "approach_recede": 42464,
    "linear": 42645,
    "circular": 42160
  },
  "split_counts": {
    "test": 6436,
    "train": 248102
  },
  "audio_format": "PCM16 WAV",
  "sample_rate": 16000,
  "channels": 4,
  "waypoint_rate_hz": 10,
  "waypoints_per_10s_clip": 101,
  "renderer_version": "physwave_foa_release_v1_pcm16",
  "metadata_path": "/media/volume/Spatial_audio_dataset_upload/metadata.jsonl",
  "metadata_shards_dir": "/media/volume/Spatial_audio_dataset_upload/metadata_shards",
  "parquet_rows": 254538,
  "parquet_split_counts": {
    "test": 6436,
    "train": 248102
  },
  "parquet_shard_counts": {
    "test": 7,
    "train": 249
  },
  "rows_per_shard": 1000,
  "row_group_size": 100,
  "audio_column": "audio",
  "parquet_compression": "zstd"
}
```