Data Description
Prompt Types
There are six different prompt types, encoding the task with was requested of the child. Please see section 3.1 of Eshky et al (2018) for further details.
| Type | ID | UXTD | UXSSD | UPX | UX2020 |
|---|---|---|---|---|---|
| Words | A | 962 (26) | 2708 (291) | 3838 (455) | 935 (109) |
| Non-words | B | 607 (27) | 495 (59) | 560 (60) | - |
| Sentence | C | 0 (0) | 445 (35) | 1020 (128) | 553 (15) |
| Articulatory | D | 2934 (45) | 132 (17) | 211 (21) | 574 (22) |
| Non-speech | E | 116 (2) | 9 (1) | 302 (1) | 88 (2) |
| Other | F | 0 (0) | 56 (12) | 61 (1) | - |
| Total | 4619 (100) | 3845 (415) | 5992 (682) | 2150 (148) |
Speech Hours
The following hours of speech and silence are rounded to two decimal places and are estimated using the speaker labels that accompany each dataset.
| UXTD | UXSSD | UPX | |
|---|---|---|---|
| Child speech | 2.24 (28.39%) | 3.66 (34.45%) | 7.27 (38.70%) |
| SLT speech | 1.24 (15.74%) | 1.81 (16.99%) | 1.92 (10.23%) |
| Total speech | 3.47 (44.12%) | 5.47 (51.43%) | 9.19 (48.93%) |
| Initial silence | 1.41 (17.96%) | 0.91 (8.55%) | 0.78 (4.17%) |
| Medial silence | 1.99 (25.30%) | 3.48 (32.69%) | 7.11 (37.83%) |
| Final silence | 0.99 (12.61%) | 0.78 (7.32%) | 1.70 (9.07%) |
| Total silence | 4.40 (55.88%) | 5.16 (48.57%) | 9.59 (51.07%) |
| Total Audio | 7.87 | 10.63 | 18.78 |