speaker_id stringdate 7033-01-01 00:00:00 8992-01-01 00:00:00 | name stringclasses 3
values | accent stringclasses 2
values | country_of_birth stringclasses 2
values | bio stringclasses 1
value | role stringclasses 2
values | first_recorded timestamp[s]date 2026-10-14 00:00:00 2026-10-29 00:00:00 ⌀ |
|---|---|---|---|---|---|---|
8992 | Daniel Rosehill | Irish English | Ireland | Lives in Jerusalem. | Project creator; main speaker | 2026-10-14T00:00:00 |
8137 | Hannah | US English | United States | Lives in Jerusalem. | Contributor | 2026-10-29T00:00:00 |
7033 | Mike | null | null | null | Contributor | null |
Hebrish Speakers
A small dataset to provide accreditation of all contributors to the Hebrish project: one row per person whose
voice is in its recordings. Each speaker has a permanent speaker_id, which maps them across every Hebrish dataset.
| Column | What it holds |
|---|---|
speaker_id |
Permanent contributor ID: a randomly drawn 4-digit number, never reused. Every Hebrish recording dataset has a speaker_id column with the same value |
name |
Name as it appears in the datasets' speaker column |
accent |
Accent of English |
country_of_birth |
Country of birth |
bio |
Short biography |
role |
Part in the project |
first_recorded |
Date of the speaker's first recording |
Speaker labels in the recording datasets are checked against each speaker's voice (ECAPA speaker embeddings) before publishing; see hebrish.app/dataset#integrity. The voiceprints themselves are kept private.
Want to contribute recordings? See hebrish.app/dataset.
- Downloads last month
- 265