Dataset Viewer
Auto-converted to Parquet Duplicate
id
stringlengths
14
16
speaker_id
stringclasses
8 values
chapter_id
stringclasses
19 values
duration_s
float64
1.87
33.9
reference
stringlengths
5
576
transcript
stringlengths
7
401
reference_normalized
stringlengths
5
576
transcript_normalized
stringlengths
7
393
wer
float64
0
1
embedding
listlengths
384
384
1089-134686-0000
1089
134686
10.435
HE HOPED THERE WOULD BE STEW FOR DINNER TURNIPS AND CARROTS AND BRUISED POTATOES AND FAT MUTTON PIECES TO BE LADLED OUT IN THICK PEPPERED FLOUR FATTENED SAUCE
He hoped there would be stew for dinner, turnips and carrots and bruised potatoes and fat mutton pieces to be ladled out in thick peppered flour-fattened sauce.
he hoped there would be stew for dinner turnips and carrots and bruised potatoes and fat mutton pieces to be ladled out in thick peppered flour fattened sauce
he hoped there would be stew for dinner turnips and carrots and bruised potatoes and fat mutton pieces to be ladled out in thick peppered flour fattened sauce
0
[ -0.009833202697336674, 0.05629004165530205, -0.03832649067044258, 0.06954573094844818, -0.03370806947350502, -0.022210337221622467, 0.016130493953824043, -0.011243053711950779, -0.03533947467803955, -0.06257369369268417, 0.03188065066933632, -0.08925160020589828, -0.020016541704535484, -0....
1089-134686-0001
1089
134686
3.275
STUFF IT INTO YOU HIS BELLY COUNSELLED HIM
Stuff it into you, his belly counseled him.
stuff it into you his belly counseled him
stuff it into you his belly counseled him
0
[ 0.06028902903199196, 0.1135810911655426, -0.038239486515522, 0.0446050688624382, 0.029489759355783463, -0.03351138159632683, 0.12069651484489441, -0.016000736504793167, -0.002779534086585045, -0.0851462259888649, -0.09282436966896057, 0.0001373728591715917, -0.031632933765649796, 0.0950913...
1089-134686-0002
1089
134686
6.625
AFTER EARLY NIGHTFALL THE YELLOW LAMPS WOULD LIGHT UP HERE AND THERE THE SQUALID QUARTER OF THE BROTHELS
After early nightfall the yellow lamps would light up here and there the squalid quarter of the brothels.
after early nightfall the yellow lamps would light up here and there the squalid quarter of the brothels
after early nightfall the yellow lamps would light up here and there the squalid quarter of the brothels
0
[ 0.017417477443814278, 0.05966128408908844, -0.02455119602382183, 0.05748237296938896, -0.03881083428859711, 0.00926385447382927, 0.017934197559952736, -0.0856260284781456, -0.04068142548203468, 0.03237804397940636, 0.08219577372074127, -0.02270500920712948, 0.029086092486977577, 0.04184390...
1089-134686-0003
1089
134686
2.68
HELLO BERTIE ANY GOOD IN YOUR MIND
Hello Bertie, any good in your mind?
hello bertie any good in your mind
hello bertie any good in your mind
0
[ -0.007012072950601578, -0.009036519564688206, 0.05507567897439003, 0.007505863439291716, -0.08873744308948517, 0.01206172350794077, 0.11494966596364975, -0.07427044212818146, 0.008594183251261711, -0.042416200041770935, 0.018429342657327652, 0.008414589799940586, -0.010511993430554867, -0....
1089-134686-0004
1089
134686
5.215
NUMBER TEN FRESH NELLY IS WAITING ON YOU GOOD NIGHT HUSBAND
Number ten, Fresh Nellie is waiting on you. Good night, husband.
number 10 fresh nelly is waiting on you good night husband
number 10 fresh nellie is waiting on you good night husband
0.090909
[ -0.10830686241388321, -0.09427382797002792, 0.05796719342470169, 0.013147138990461826, 0.010757953859865665, 0.10091318190097809, 0.1283215433359146, -0.029583081603050232, -0.0036734198220074177, -0.061805423349142075, -0.03673412650823593, 0.017892247065901756, -0.026211412623524666, 0.0...
1089-134686-0005
1089
134686
9.635
THE MUSIC CAME NEARER AND HE RECALLED THE WORDS THE WORDS OF SHELLEY'S FRAGMENT UPON THE MOON WANDERING COMPANIONLESS PALE FOR WEARINESS
The music came nearer and he recalled the words, the words of Shelley's fragment upon the moon wandering companionless, pale for weariness.
the music came nearer and he recalled the words the words of shelley is fragment upon the moon wandering companionless pale for weariness
the music came nearer and he recalled the words the words of shelley is fragment upon the moon wandering companionless pale for weariness
0
[ 0.018519248813390732, 0.008049064315855503, 0.11900067329406738, 0.058161087334156036, 0.04499858245253563, 0.07440101355314255, 0.11516871303319931, 0.00788835808634758, -0.06978932023048401, -0.07356671988964081, 0.010503498837351799, 0.006105409003794193, 0.01630108430981636, -0.0820225...
1089-134686-0006
1089
134686
10.555
THE DULL LIGHT FELL MORE FAINTLY UPON THE PAGE WHEREON ANOTHER EQUATION BEGAN TO UNFOLD ITSELF SLOWLY AND TO SPREAD ABROAD ITS WIDENING TAIL
The dull light fell more faintly upon the page whereon another equation began to unfold itself slowly and to spread abroad its widening tail.
the dull light fell more faintly upon the page whereon another equation began to unfold itself slowly and to spread abroad its widening tail
the dull light fell more faintly upon the page whereon another equation began to unfold itself slowly and to spread abroad its widening tail
0
[ -0.06706544756889343, -0.03413723036646843, 0.10046722739934921, 0.15287335216999054, 0.09400849044322968, -0.05099596828222275, -0.02126937359571457, 0.05655449628829956, 0.0475073978304863, 0.025455614551901817, 0.10801167041063309, 0.040817875415086746, -0.06342468410730362, 0.066000081...
1089-134686-0007
1089
134686
4.275
A COLD LUCID INDIFFERENCE REIGNED IN HIS SOUL
A cold, lucid indifference reigned in his soul.
a cold lucid indifference reigned in his soul
a cold lucid indifference reigned in his soul
0
[ 0.010031787678599358, 0.08718034625053406, 0.04916737228631973, 0.10993140935897827, -0.016574330627918243, 0.04840080067515373, 0.14504320919513702, 0.023284584283828735, 0.046193674206733704, -0.022634729743003845, 0.026077033951878548, -0.00743474904447794, 0.022892199456691742, -0.0283...
1089-134686-0008
1089
134686
6.73
THE CHAOS IN WHICH HIS ARDOUR EXTINGUISHED ITSELF WAS A COLD INDIFFERENT KNOWLEDGE OF HIMSELF
The chaos in which his ardor extinguished itself was a cold indifferent knowledge of
the chaos in which his ardor extinguished itself was a cold indifferent knowledge of himself
the chaos in which his ardor extinguished itself was a cold indifferent knowledge of
0.066667
[ -0.005464439745992422, 0.04325088486075401, 0.02924024686217308, 0.10371583700180054, -0.008461974561214447, -0.03134320676326752, 0.06435989588499069, -0.02906973846256733, -0.005134947132319212, -0.007786339148879051, -0.0045624785125255585, 0.028622694313526154, 0.10386139899492264, -0....
1089-134686-0009
1089
134686
10.575
AT MOST BY AN ALMS GIVEN TO A BEGGAR WHOSE BLESSING HE FLED FROM HE MIGHT HOPE WEARILY TO WIN FOR HIMSELF SOME MEASURE OF ACTUAL GRACE
At most, by an alms given to a beggar whose blessing he fled from, he might hope wearily to win for himself some measure of actual grace.
at most by an alms given to a beggar whose blessing he fled from he might hope wearily to win for himself some measure of actual grace
at most by an alms given to a beggar whose blessing he fled from he might hope wearily to win for himself some measure of actual grace
0
[ 0.0126223498955369, 0.12190023064613342, -0.002569844014942646, -0.005192661192268133, -0.06446927785873413, 0.016716131940484047, 0.1480802297592163, -0.031516171991825104, -0.02233124151825905, -0.003224118845537305, 0.024437977001070976, -0.05035461112856865, 0.09930171072483063, -0.055...
1089-134686-0010
1089
134686
4.405
WELL NOW ENNIS I DECLARE YOU HAVE A HEAD AND SO HAS MY STICK
Well, now, Ennis, I declare you have a head and so has my stick.
well now ennis i declare you have a head and so has my stick
well now ennis i declare you have a head and so has my stick
0
[ -0.0035861486103385687, 0.014342746697366238, 0.007248342037200928, -0.08783483505249023, 0.05523998662829399, -0.05881081894040108, 0.15426743030548096, 0.01873697154223919, -0.011226681992411613, 0.0781126320362091, -0.03661135956645012, -0.11226164549589157, 0.008180791512131691, -0.040...
1089-134686-0011
1089
134686
12.445
ON SATURDAY MORNINGS WHEN THE SODALITY MET IN THE CHAPEL TO RECITE THE LITTLE OFFICE HIS PLACE WAS A CUSHIONED KNEELING DESK AT THE RIGHT OF THE ALTAR FROM WHICH HE LED HIS WING OF BOYS THROUGH THE RESPONSES
On Saturday mornings, when the Sodality met in the chapel to recite the little office, his place was a cushioned kneeling desk at the right of the altar from which he led his wing of boys through the responses.
on saturday mornings when the sodality met in the chapel to recite the little office his place was a cushioned kneeling desk at the right of the altar from which he led his wing of boys through the responses
on saturday mornings when the sodality met in the chapel to recite the little office his place was a cushioned kneeling desk at the right of the altar from which he led his wing of boys through the responses
0
[ 0.014339380897581577, 0.11773652583360672, -0.024118592962622643, -0.006503922864794731, -0.044724322855472565, 0.11195126175880432, 0.05389402434229851, 0.0070189787074923515, 0.08042453974485397, -0.0630154013633728, -0.05353301763534546, 0.0026962703559547663, -0.055948030203580856, 0.0...
1089-134686-0012
1089
134686
11.64
HER EYES SEEMED TO REGARD HIM WITH MILD PITY HER HOLINESS A STRANGE LIGHT GLOWING FAINTLY UPON HER FRAIL FLESH DID NOT HUMILIATE THE SINNER WHO APPROACHED HER
Her eyes seemed to regard him with mild pity. Her holiness, a strange light glowing faintly upon her frail flesh, did not humiliate the sinner who approached her.
her eyes seemed to regard him with mild pity her holiness a strange light glowing faintly upon her frail flesh did not humiliate the sinner who approached her
her eyes seemed to regard him with mild pity her holiness a strange light glowing faintly upon her frail flesh did not humiliate the sinner who approached her
0
[ 0.008935058489441872, 0.12106677889823914, 0.02147950418293476, 0.06524565070867538, 0.0696798786520958, -0.011992519721388817, 0.06209016963839531, -0.01126437820494175, -0.021528080105781555, -0.04370996356010437, 0.031439945101737976, -0.10200661420822144, -0.015870947390794754, -0.0646...
1089-134686-0013
1089
134686
7.915
IF EVER HE WAS IMPELLED TO CAST SIN FROM HIM AND TO REPENT THE IMPULSE THAT MOVED HIM WAS THE WISH TO BE HER KNIGHT
If ever he was impelled to cast sin from him and to repent, the impulse that moved him was the wish to be her knight.
if ever he was impelled to cast sin from him and to repent the impulse that moved him was the wish to be her knight
if ever he was impelled to cast sin from him and to repent the impulse that moved him was the wish to be her knight
0
[ -0.02428368479013443, 0.06449158489704132, -0.03864084184169769, 0.05008046329021454, 0.04639030620455742, 0.015290195122361183, 0.0776904746890068, -0.01753106340765953, 0.05406617373228073, -0.024994438514113426, 0.041029367595911026, -0.03339700400829315, 0.023826124146580696, -0.036791...
1089-134686-0014
1089
134686
2.225
HE TRIED TO THINK HOW IT COULD BE
He tried to think how it could be.
he tried to think how it could be
he tried to think how it could be
0
[ 0.020312437787652016, 0.036806173622608185, 0.026354363188147545, 0.006963589694350958, -0.007045947480946779, -0.06476844847202301, 0.1084001362323761, 0.01834113523364067, -0.04323185235261917, -0.0038521301466971636, -0.04794391617178917, 0.02172745019197464, -0.0032963827252388, 0.0366...
1089-134686-0015
1089
134686
5.815
BUT THE DUSK DEEPENING IN THE SCHOOLROOM COVERED OVER HIS THOUGHTS THE BELL RANG
but the dusk, deepening in the schoolroom, covered over his thoughts. The bell rang.
but the dusk deepening in the schoolroom covered over his thoughts the bell rang
but the dusk deepening in the schoolroom covered over his thoughts the bell rang
0
[ 0.07112587988376617, 0.080098956823349, 0.06445242464542389, 0.03351302817463875, -0.018173223361372948, -0.024296464398503304, 0.11339140683412552, -0.03844079002737999, 0.0896698459982872, -0.04215076565742493, 0.05659061297774315, 0.0946943387389183, -0.0434473492205143, 0.0254335105419...
1089-134686-0016
1089
134686
3.54
THEN YOU CAN ASK HIM QUESTIONS ON THE CATECHISM DEDALUS
Then you can ask him questions on the Catechism Daedalus.
then you can ask him questions on the catechism dedalus
then you can ask him questions on the catechism daedalus
0.1
[ 0.03823792561888695, 0.0635438859462738, -0.03153597563505173, 0.002273555612191558, -0.07359161972999573, -0.01995871216058731, -0.014039148576557636, -0.040891215205192566, -0.04948681965470314, -0.06919240951538086, -0.05410323664546013, 0.007757414132356644, -0.12893405556678772, -0.02...
1089-134686-0017
1089
134686
8.87
STEPHEN LEANING BACK AND DRAWING IDLY ON HIS SCRIBBLER LISTENED TO THE TALK ABOUT HIM WHICH HERON CHECKED FROM TIME TO TIME BY SAYING
Stephen, leaning back and drawing idly on his scribbler, listened to the talk about him which Harren checked from time to time by saying,
stephen leaning back and drawing idly on his scribbler listened to the talk about him which heron checked from time to time by saying
stephen leaning back and drawing idly on his scribbler listened to the talk about him which harren checked from time to time by saying
0.041667
[ -0.015337894670665264, 0.01931145414710045, -0.044389508664608, -0.007925497367978096, 0.033095795661211014, 0.04639044404029846, 0.1628422886133194, 0.0029509514570236206, -0.006670096423476934, -0.03683467581868172, -0.08774010092020035, -0.012471772730350494, -0.018316561356186867, -0.0...
1089-134686-0018
1089
134686
15.72
IT WAS STRANGE TOO THAT HE FOUND AN ARID PLEASURE IN FOLLOWING UP TO THE END THE RIGID LINES OF THE DOCTRINES OF THE CHURCH AND PENETRATING INTO OBSCURE SILENCES ONLY TO HEAR AND FEEL THE MORE DEEPLY HIS OWN CONDEMNATION
It was strange, too, that he found an arid pleasure in following up to the end the rigid lines of the doctrines of the church, and penetrating into obscure silences only to hear and feel the more deeply his own condemnation.
it was strange too that he found an arid pleasure in following up to the end the rigid lines of the doctrines of the church and penetrating into obscure silences only to hear and feel the more deeply his own condemnation
it was strange too that he found an arid pleasure in following up to the end the rigid lines of the doctrines of the church and penetrating into obscure silences only to hear and feel the more deeply his own condemnation
0
[ 0.06959183514118195, 0.04817936196923256, 0.02677266113460064, 0.04623989388346672, 0.04391583427786827, -0.01185834500938654, 0.08621763437986374, -0.033822428435087204, 0.010269500315189362, -0.06165522336959839, -0.03620956838130951, -0.015430162660777569, 0.024305904284119606, -0.03723...
1089-134686-0019
1089
134686
13.895
THE SENTENCE OF SAINT JAMES WHICH SAYS THAT HE WHO OFFENDS AGAINST ONE COMMANDMENT BECOMES GUILTY OF ALL HAD SEEMED TO HIM FIRST A SWOLLEN PHRASE UNTIL HE HAD BEGUN TO GROPE IN THE DARKNESS OF HIS OWN STATE
The sentence of St. James, which says that he who offends against one commandment becomes guilty of all, had seemed to him first a swollen phrase, until he had begun to grope in the darkness of his own state.
the sentence of saint james which says that he who offends against one commandment becomes guilty of all had seemed to him 1st a swollen phrase until he had begun to grope in the darkness of his own state
the sentence of saint james which says that he who offends against one commandment becomes guilty of all had seemed to him 1st a swollen phrase until he had begun to grope in the darkness of his own state
0
[ -0.01579534448683262, 0.0709538534283638, 0.05316263064742088, 0.06734779477119446, 0.034667253494262695, 0.02668657898902893, 0.11896488070487976, -0.009990300983190536, -0.018515055999159813, -0.022153321653604507, 0.03631694242358208, -0.005930256098508835, 0.03915996104478836, -0.01975...
1089-134686-0020
1089
134686
16.79
IF A MAN HAD STOLEN A POUND IN HIS YOUTH AND HAD USED THAT POUND TO AMASS A HUGE FORTUNE HOW MUCH WAS HE OBLIGED TO GIVE BACK THE POUND HE HAD STOLEN ONLY OR THE POUND TOGETHER WITH THE COMPOUND INTEREST ACCRUING UPON IT OR ALL HIS HUGE FORTUNE
If a man had stolen a pound in his youth, and had used that pound to amass a huge fortune, how much was he obliged to give back? The pound he had stolen only, or the pound together with the compound interest accruing upon it, or all his huge fortune?
if a man had stolen a pound in his youth and had used that pound to amass a huge fortune how much was he obliged to give back the pound he had stolen only or the pound together with the compound interest accruing upon it or all his huge fortune
if a man had stolen a pound in his youth and had used that pound to amass a huge fortune how much was he obliged to give back the pound he had stolen only or the pound together with the compound interest accruing upon it or all his huge fortune
0
[ 0.017169121652841568, 0.15243205428123474, -0.03374496474862099, -0.0013140045339241624, -0.05931913107633591, -0.02990785241127014, 0.04987437278032303, 0.02766914665699005, 0.03027883917093277, 0.04769390448927879, 0.059864576905965805, -0.016493001952767372, 0.00990572664886713, 0.00219...
1089-134686-0021
1089
134686
6.55
IF A LAYMAN IN GIVING BAPTISM POUR THE WATER BEFORE SAYING THE WORDS IS THE CHILD BAPTIZED
If a layman, in giving baptism, pour the water before saying the words, is the child baptized?
if a layman in giving baptism pour the water before saying the words is the child baptized
if a layman in giving baptism pour the water before saying the words is the child baptized
0
[ 0.05387168377637863, 0.07845360785722733, -0.026886271312832832, -0.020758045837283134, -0.09511980414390564, -0.035766150802373886, 0.10557437688112259, -0.10436628758907318, 0.012111077085137367, -0.05098965764045715, 0.041954901069402695, -0.02977241389453411, -0.07473187893629074, 0.07...
End of preview. Expand in Data Studio

LibriSpeech test-clean: whisper-medium.en batch transcripts (first 500 utterances)

This is a real batch transcription job from Boole Inference (Boole AI, Inc.). It ran on CPU, and every number below was measured in this run.

Results

Model openai/whisper-medium.en (Apache-2.0), fp32, greedy decoding
Files processed 500 / 500
Failed / retried 0 failed, 0 retried batches
Audio transcribed 3,809.0 s (63.5 min); utterances range from 1.9 s to 33.9 s
Wall time for the transcription job 1,795 s (29.9 min), including model load; the transcription loop alone took 1,788.9 s
Speed about 2.1× real time (63.5 audio-min in 29.9 wall-min)
WER (normalized) 2.29% = 244 errors / 10,634 reference words (126 substitutions, 105 deletions, 13 insertions)
Transcript embeddings all-MiniLM-L6-v2, 384 dimensions, 0.84 s for 500 rows

Hardware: CPU only, 8-core Intel Xeon VM, 15 GB RAM, no GPU. Batch size 8, 8 threads.

Subset

  • Source: LibriSpeech test-clean (Panayotov et al., 2015), license CC BY 4.0, downloaded from OpenSLR.
  • Subset: the first 500 utterances of test-clean (2,620 total) after sorting by utterance ID. That runs from 1089-134686-0000 to 1995-1837-0018 and covers 8 speakers.
  • Because only 8 speakers are included, this is not a substitute for a full test-clean benchmark.

WER method

  • Computed as corpus-level WER with jiwer.process_words over all 500 pairs.
  • Both the reference and the hypothesis are normalized first with Whisper's EnglishTextNormalizer, via WhisperTokenizer.normalize. This is the standard Whisper evaluation normalizer: it lowercases, removes punctuation, and standardizes spelling, numbers and contractions.
  • For transparency: lowercase-only WER without that normalization is 16.48%, mostly due to punctuation, because LibriSpeech references have none.
  • Each row also has a per-utterance wer.

Columns

  • id, speaker_id, chapter_id, duration_s
  • reference: the original LibriSpeech text
  • transcript: the raw Whisper output
  • reference_normalized, transcript_normalized, wer
  • embedding: all-MiniLM-L6-v2 over transcript, L2-normalized

Scripts

scripts/transcribe.py takes the model, N and an output prefix; it resumes from a partial JSONL and retries a failed batch once. scripts/score_and_embed.py computes WER and embeddings. Raw stats are in run_stats.json.

Limitations

  • Read audiobook speech in clean conditions is easier than calls or meetings, so expect higher WER on real-world audio.
  • The subset is the first 500 utterances by ID, not a random sample.
  • One long utterance ended early (per-utterance WER 0.80), and proper names make up some errors (for example "Stephanos Dedalos").
  • Runtime reflects this specific CPU VM. GPU runs would be much faster.
Downloads last month
33