Voiceprint Comparison: get task
GET/api/technology/speaker-identification-voiceprint-comparison/:task_id
Retrieve task for comparison of voiceprints.
How a Voiceprint Comparison result is arranged
See Voiceprint Comparison task request documentation for details on how the request is arranged.
The result of a Voiceprint Comparison task is a matrix of scores represented as a single vector in row-major order, expressed as follows:
[
vp_a_1 x vp_b_1, vp_a_2 x vp_b_1, ..., vp_a_M x vp_b_1,
vp_a_1 x vp_b_2, vp_a_2 x vp_b_2, ..., vp_a_M x vp_b_2,
...,
vp_a_1 x vp_b_N, vp_a_2 x vp_b_N, ..., vp_a_M x vp_b_N
]
where M = size of voiceprints_a list (columns_count) and
N = size of voiceprints_b list (rows_count). In other words, going through the list of
scores is the equivalent of traversing the matrix row by row.
The result also contains the number of matrix rows and columns. The columns_count and
rows_count correspond to the comparison's input request parameters voiceprints_a and
voiceprints_b respectively.
For example, the following result data in the response:
{
"scores": {
"columns_count": 3,
"rows_count": 3,
"values": [
3, 1, 2,
8, 2, -6,
11, 2, 7
]
}
}
represent the following result matrix:
| voiceprints_a list | ||||
|---|---|---|---|---|
| vp_a_1 | vp_a_2 | vp_a_3 | ||
| voiceprints_b list | vp_b_1 | 3 | 1 | 2 |
| vp_b_2 | 8 | 2 | -6 | |
| vp_b_3 | 11 | 2 | 7 | |
Request
Responses
- 200
- 400
- 404
- 405
Voiceprint Comparison task.
The response body contains task info in the task field. Depending on the task processing outcome, either result or error field is populated.
Request payload data was invalid and could not be parsed.
The requested task does not exist.
The provided task_id may be wrong or the task may have expired already.
Invalid HTTP method.