Datasets:
audio audioduration (s) 7.77 11.4 | id stringclasses 5
values | text stringclasses 5
values | text_he stringclasses 5
values | duration_s float64 7.77 11.4 | categories listlengths 1 1 | target_terms listlengths 0 0 | prompt_id stringclasses 5
values | prompt_source stringclasses 1
value | prompt_origin stringclasses 1
value | mode stringclasses 1
value | edited bool 1
class | device stringclasses 1
value | microphone stringclasses 1
value | environment stringclasses 1
value | conditions listlengths 0 0 | noise_level stringclasses 0
values | background_speech_languages listlengths 0 0 | noise_flags listlengths 0 0 | clipped bool 1
class | notes stringclasses 0
values | hebrew_pronunciation stringclasses 0
values | variant_groups listlengths 0 0 | variant_spelling stringclasses 0
values | task stringclasses 1
value | spoken_text stringclasses 5
values | aside_text stringclasses 5
values | aside_position stringclasses 3
values | verbatim_text stringclasses 0
values | prompt_categories listlengths | source_prompt_id stringclasses 0
values | prompt_kind stringclasses 0
values | prompt_note stringclasses 0
values | term_language stringclasses 0
values | paragraph_text stringclasses 0
values | model_hypothesis stringclasses 0
values | asr_model stringclasses 0
values | speaker stringclasses 1
value | speaker_id stringdate 8992-01-01 00:00:00 8992-01-01 00:00:00 | contributor stringclasses 1
value | speaker_gender stringclasses 1
value | sample_rate int64 16k 48k | recorded_at stringclasses 5
values | priority_word stringclasses 0
values | bit_depth int64 16 24 | audio_format stringclasses 2
values | native_rate int64 48k 48k ⌀ | capture_mode stringclasses 1
value | tts_eligible bool 1
class | tts_split stringclasses 0
values | transcript_status stringclasses 1
value | prompt_set stringclasses 0
values | topic stringclasses 0
values | style stringclasses 0
values | style_instruction stringclasses 0
values | written_form_uncertain bool 0
classes | session_id stringclasses 1
value | room_tone_id stringclasses 0
values | room stringclasses 0
values | location stringclasses 0
values | location_name stringclasses 0
values | location_note stringclasses 0
values | latitude float64 | longitude float64 | location_accuracy_m float64 | lead_trim_s int64 0 0 ⌀ | peak_dbfs float64 -5.6 -4.7 ⌀ | floor_dbfs float64 -57.3 -55.6 ⌀ | speech_dbfs float64 -27.9 -20.8 ⌀ | snr_db float64 27.7 36.5 ⌀ | clipped_samples int64 0 0 ⌀ | edge_head_dbfs float64 -44.5 -40.4 ⌀ | edge_tail_dbfs float64 -47.1 -43.6 ⌀ | browser_processing bool 1
class | sentence_id stringclasses 5
values | pronunciation stringclasses 1
value | extra stringclasses 0
values |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
20261021T114900Z_plhufv | We need to buy a new washing machine before the old one breaks down completely, ideally under $600. | We need to buy a new washing machine before the old one breaks down completely, ideally under $600. | 11.43 | [
"asides"
] | [] | asides_004970ebeb | generated:claude-self-written | generated | bank | false | OnePlus Nord 3 5G | internal | quiet-indoor | [] | null | [] | [] | false | null | null | [] | null | asides | We need to buy a new washing machine — Ezra, not now, sweetie — before the old one breaks down completely, ideally under $600. | Ezra, not now, sweetie | middle | null | null | null | null | null | null | null | null | null | Daniel Rosehill | 8992 | Daniel Rosehill | male | 16,000 | 2026-10-21T11:49:00.273Z | null | 16 | wav | null | null | null | null | script | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | p_ce6730e054 | [] | null | |
20261021T120048Z_nmecyo | Thanks again for your help with the move. I owe you a coffee, and maybe lunch as well. | Thanks again for your help with the move. I owe you a coffee, and maybe lunch as well. | 8.142 | [
"asides"
] | [] | asides_01f869b8c0 | generated:claude-self-written | generated | bank | false | OnePlus Nord 3 5G | internal | quiet-indoor | [] | null | [] | [] | false | null | null | [] | null | asides | Hey Sam, just a minute. Thanks again for your help with the move. I owe you a coffee, and maybe lunch as well. | Hey Sam, just a minute. | start | null | null | null | null | null | null | null | null | null | Daniel Rosehill | 8992 | Daniel Rosehill | male | 16,000 | 2026-10-21T12:00:48.913Z | null | 16 | wav | null | null | null | null | script | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | null | p_19ea495275 | [] | null | |
20261027T173314Z_5yk1c7 | I have finished the draft and will send it to you tonight for comments. | I have finished the draft and will send it to you tonight for comments. | 7.767 | [
"asides"
] | [] | asides_24d4f51789 | generated:claude-self-written | generated | bank | false | OnePlus Nord 3 5G | internal | quiet-indoor | [] | null | [] | [] | false | null | null | [] | null | asides | I have finished the draft and will send it to you tonight for comments. Perfect timing, thanks. | Perfect timing, thanks. | end | null | null | null | null | null | null | null | null | null | Daniel Rosehill | 8992 | Daniel Rosehill | male | 48,000 | 2026-10-27T17:33:14.614Z | null | 24 | flac | 48,000 | stt | false | null | script | null | null | null | null | null | 20261027T173244Z_nn77hu | null | null | null | null | null | null | null | null | 0 | -5.6 | -55.6 | -27.9 | 27.7 | 0 | -40.4 | -46.7 | false | p_76695e7309 | [] | null | |
20261027T173325Z_xuso6d | The plumber is coming at 10:30, so someone needs to be home to let him in. | The plumber is coming at 10:30, so someone needs to be home to let him in. | 9.631 | [
"asides"
] | [] | asides_f63415159c | generated:claude-self-written | generated | bank | false | OnePlus Nord 3 5G | internal | quiet-indoor | [] | null | [] | [] | false | null | null | [] | null | asides | The plumber is coming at 10:30, so someone needs to be home to let him in. You can leave it by the gate. | You can leave it by the gate. | end | null | null | null | null | null | null | null | null | null | Daniel Rosehill | 8992 | Daniel Rosehill | male | 48,000 | 2026-10-27T17:33:25.269Z | null | 24 | flac | 48,000 | stt | false | null | script | null | null | null | null | null | 20261027T173244Z_nn77hu | null | null | null | null | null | null | null | null | 0 | -5.1 | -56.5 | -21.6 | 34.9 | 0 | -44.5 | -43.6 | false | p_372cf028eb | [] | null | |
20261027T173335Z_cnzw36 | Hi Jenny, the keys are with the neighbour, and the code for the gate is on the fridge. | Hi Jenny, the keys are with the neighbour, and the code for the gate is on the fridge. | 8.722 | [
"asides"
] | [] | asides_e4f1a3ecb4 | generated:claude-self-written | generated | bank | false | OnePlus Nord 3 5G | internal | quiet-indoor | [] | null | [] | [] | false | null | null | [] | null | asides | Hi Jenny, the keys are with the neighbour, and the code for the gate is on the fridge. Please take your seat, Jacob. | Please take your seat, Jacob. | end | null | null | null | null | null | null | null | null | null | Daniel Rosehill | 8992 | Daniel Rosehill | male | 48,000 | 2026-10-27T17:33:35.070Z | null | 24 | flac | 48,000 | stt | false | null | script | null | null | null | null | null | 20261027T173244Z_nn77hu | null | null | null | null | null | null | null | null | 0 | -4.7 | -57.3 | -20.8 | 36.5 | 0 | -43.8 | -47.1 | false | p_25b1198776 | [] | null |
Spoken Aside Sentences
Audio dataset, split-out. Rows copied from the master, danielrosehill/Read-Aloud-Voice-Sentences, so one task can be browsed alone. Training reads the master, not this copy. Every audio dataset as one tree, with hours: the audio index. How every dataset fits together: the dataset index.
One speaker dictating a sentence and, partway through, saying something to someone in the room ("come in, Hannah"). text excludes the aside; aside_text and aside_position (start, middle, end) give it; spoken_text is the full script.
Website: hebrish.app: get the Hebrish keyboard for Android and the desktop apps, see the results, and find every model, dataset and demo in one place.
Two tabs. sentences lists every ground-truth sentence with a takes count, so
takes = 0 shows what has not been recorded. takes has one row per recording with its audio, joined on sentence_id; a
sentence can have several takes. load_dataset("Hebrish/Spoken-Aside-Sentences", "sentences") reads the text; the default config reads the takes.
Takes come from danielrosehill/Read-Aloud-Voice-Sentences, the full dataset, with the same ids and
train/test split, by scripts/sync_subsets.py in the project repository. Do not edit this dataset directly. New takes append and
only the sentence counts are rewritten.
Speaker: Daniel Rosehill, the dataset's author, who was born in Ireland and speaks English with an Irish accent, and who lives in Jerusalem. Every recording is his voice. Deliberate variation: he sometimes changes his speaking speed on purpose (fast and rushed, or slow) and sometimes says a word the way people commonly mispronounce it, to make the models robust to real speech. The transcript is always the sentence as written, so a take whose speed or pronunciation sounds off against its card is usually intentional, not a recording error.
All audio was recorded on a OnePlus Nord 3 5G phone using its internal microphone, with browser echo cancellation, noise suppression and auto gain turned off.
Layout
sentences/train.parquet one row per sentence: `id`, `text`, `takes`, `prompt_ids`, `served` and the sentence columns
data/train-*.parquet one row per take: `audio` (the FLAC or WAV file, embedded) and every column
meta/train.jsonl index: the same rows without audio, plus `shard`, `audio_path` and the audio's byte range
meta/schema.json column types
data/test-*.parquet, meta/test.jsonl held-out set, never trained on
Audio is 48 kHz mono 24-bit FLAC for rows from 2026-10-22 and 16 kHz mono 16-bit WAV before.
The test split is term-disjoint: a held-out term never appears in train.
load_dataset("Hebrish/Spoken-Aside-Sentences") works directly.
Columns
A take's sentence_id is its row's id in sentences. A sentence with served: false is no longer offered by the app; it is kept because a take used it. The columns are the same as in the main dataset: text (ground truth, punctuated and cased like
Parakeet v2 output), text_he (the same transcript with its Hebrew terms in Hebrew script, for
Hebrew-capable models), categories, target_terms, prompt_source, prompt_origin (manual / dataset / generated), edited, duration_s,
environment, conditions, device, speaker, speaker_gender, contributor, hebrew_pronunciation, variant_groups, variant_spelling and recorded_at. The main dataset's card describes them.
Licence
CC-BY-4.0, with sentence-text attribution per prompt_source as described in the
main dataset's licence section.
- Downloads last month
- 246