Datasets:

WAZOBIALABS
/

nigerian-pidgin-voice-text

Error code:   DatasetGenerationCastError
Exception:    DatasetGenerationCastError
Message:      An error occurred while generating the dataset

All the data files must have the same columns, but at some point there are 3 new columns ({'field_name', 'value', 'definition'}) and 15 missing columns ({'prosody_match', 'register', 'emotion_category', 'entry_id', 'date_added', 'emotion_secondary', 'pidgin_text', 'sarcasm_flag', 'intensity', 'speaker_gender', 'topic_domain', 'annotator_notes', 'source_type', 'sentiment', 'english_gloss'}).

This happened while the csv dataset builder was generating data using

hf://datasets/WAZOBIALABS/nigerian-pidgin-voice-text/wazobia_codebook_v07.csv (at revision 3a3bca74f49e4daec1d3cff3359209c15033b0cd), [/tmp/hf-datasets-cache/medium/datasets/94122100367347-config-parquet-and-info-WAZOBIALABS-nigerian-pidg-6644c64e/hub/datasets--WAZOBIALABS--nigerian-pidgin-voice-text/snapshots/3a3bca74f49e4daec1d3cff3359209c15033b0cd/Wazobia Labs — Nigerian Pidgin Dataset v0.6 - Sheet1.csv (origin=hf://datasets/WAZOBIALABS/nigerian-pidgin-voice-text@3a3bca74f49e4daec1d3cff3359209c15033b0cd/Wazobia Labs — Nigerian Pidgin Dataset v0.6 - Sheet1.csv), /tmp/hf-datasets-cache/medium/datasets/94122100367347-config-parquet-and-info-WAZOBIALABS-nigerian-pidg-6644c64e/hub/datasets--WAZOBIALABS--nigerian-pidgin-voice-text/snapshots/3a3bca74f49e4daec1d3cff3359209c15033b0cd/wazobia_codebook_v07.csv (origin=hf://datasets/WAZOBIALABS/nigerian-pidgin-voice-text@3a3bca74f49e4daec1d3cff3359209c15033b0cd/wazobia_codebook_v07.csv), /tmp/hf-datasets-cache/medium/datasets/94122100367347-config-parquet-and-info-WAZOBIALABS-nigerian-pidg-6644c64e/hub/datasets--WAZOBIALABS--nigerian-pidgin-voice-text/snapshots/3a3bca74f49e4daec1d3cff3359209c15033b0cd/wazobia_pidgin_v07_500.csv (origin=hf://datasets/WAZOBIALABS/nigerian-pidgin-voice-text@3a3bca74f49e4daec1d3cff3359209c15033b0cd/wazobia_pidgin_v07_500.csv)]

Please either edit the data files to have matching columns, or separate them into different configurations (see docs at https://hf.co/docs/hub/datasets-manual-configuration#multiple-configurations)
Traceback:    Traceback (most recent call last):
                File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 1800, in _prepare_split_single
                  writer.write_table(table)
                File "/usr/local/lib/python3.12/site-packages/datasets/arrow_writer.py", line 765, in write_table
                  self._write_table(pa_table, writer_batch_size=writer_batch_size)
                File "/usr/local/lib/python3.12/site-packages/datasets/arrow_writer.py", line 773, in _write_table
                  pa_table = table_cast(pa_table, self._schema)
                             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/table.py", line 2321, in table_cast
                  return cast_table_to_schema(table, schema)
                         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/table.py", line 2249, in cast_table_to_schema
                  raise CastError(
              datasets.table.CastError: Couldn't cast
              field_name: string
              value: string
              definition: string
              -- schema metadata --
              pandas: '{"index_columns": [{"kind": "range", "name": null, "start": 0, "' + 610
              to
              {'entry_id': Value('string'), 'pidgin_text': Value('string'), 'english_gloss': Value('string'), 'sentiment': Value('string'), 'emotion_category': Value('string'), 'emotion_secondary': Value('string'), 'register': Value('string'), 'sarcasm_flag': Value('string'), 'prosody_match': Value('string'), 'annotator_notes': Value('string'), 'topic_domain': Value('string'), 'speaker_gender': Value('string'), 'intensity': Value('int64'), 'source_type': Value('string'), 'date_added': Value('string')}
              because column names don't match
              
              During handling of the above exception, another exception occurred:
              
              Traceback (most recent call last):
                File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 1347, in compute_config_parquet_and_info_response
                  parquet_operations = convert_to_parquet(builder)
                                       ^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 980, in convert_to_parquet
                  builder.download_and_prepare(
                File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 882, in download_and_prepare
                  self._download_and_prepare(
                File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 943, in _download_and_prepare
                  self._prepare_split(split_generator, **prepare_split_kwargs)
                File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 1646, in _prepare_split
                  for job_id, done, content in self._prepare_split_single(
                                               ^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 1802, in _prepare_split_single
                  raise DatasetGenerationCastError.from_cast_error(
              datasets.exceptions.DatasetGenerationCastError: An error occurred while generating the dataset
              
              All the data files must have the same columns, but at some point there are 3 new columns ({'field_name', 'value', 'definition'}) and 15 missing columns ({'prosody_match', 'register', 'emotion_category', 'entry_id', 'date_added', 'emotion_secondary', 'pidgin_text', 'sarcasm_flag', 'intensity', 'speaker_gender', 'topic_domain', 'annotator_notes', 'source_type', 'sentiment', 'english_gloss'}).
              
              This happened while the csv dataset builder was generating data using
              
              hf://datasets/WAZOBIALABS/nigerian-pidgin-voice-text/wazobia_codebook_v07.csv (at revision 3a3bca74f49e4daec1d3cff3359209c15033b0cd), [/tmp/hf-datasets-cache/medium/datasets/94122100367347-config-parquet-and-info-WAZOBIALABS-nigerian-pidg-6644c64e/hub/datasets--WAZOBIALABS--nigerian-pidgin-voice-text/snapshots/3a3bca74f49e4daec1d3cff3359209c15033b0cd/Wazobia Labs — Nigerian Pidgin Dataset v0.6 - Sheet1.csv (origin=hf://datasets/WAZOBIALABS/nigerian-pidgin-voice-text@3a3bca74f49e4daec1d3cff3359209c15033b0cd/Wazobia Labs — Nigerian Pidgin Dataset v0.6 - Sheet1.csv), /tmp/hf-datasets-cache/medium/datasets/94122100367347-config-parquet-and-info-WAZOBIALABS-nigerian-pidg-6644c64e/hub/datasets--WAZOBIALABS--nigerian-pidgin-voice-text/snapshots/3a3bca74f49e4daec1d3cff3359209c15033b0cd/wazobia_codebook_v07.csv (origin=hf://datasets/WAZOBIALABS/nigerian-pidgin-voice-text@3a3bca74f49e4daec1d3cff3359209c15033b0cd/wazobia_codebook_v07.csv), /tmp/hf-datasets-cache/medium/datasets/94122100367347-config-parquet-and-info-WAZOBIALABS-nigerian-pidg-6644c64e/hub/datasets--WAZOBIALABS--nigerian-pidgin-voice-text/snapshots/3a3bca74f49e4daec1d3cff3359209c15033b0cd/wazobia_pidgin_v07_500.csv (origin=hf://datasets/WAZOBIALABS/nigerian-pidgin-voice-text@3a3bca74f49e4daec1d3cff3359209c15033b0cd/wazobia_pidgin_v07_500.csv)]
              
              Please either edit the data files to have matching columns, or separate them into different configurations (see docs at https://hf.co/docs/hub/datasets-manual-configuration#multiple-configurations)

Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.

entry_id string	pidgin_text string	english_gloss string	sentiment string	emotion_category string	emotion_secondary null	register string	sarcasm_flag string	prosody_match string	annotator_notes string	topic_domain string	speaker_gender string	intensity int64	source_type string	date_added string
WZ-T-0001	How you dey?	How are you?	positive	joy	null	casual	no	matched	Standard Lagos greeting	greeting	neutral	1	original	2026-04-23
WZ-T-0002	How body?	How are you feeling?	neutral	neutral	null	casual	no	matched	More intimate than how you dey	greeting	neutral	1	original	2026-04-23
WZ-T-0003	You don chop?	Have you eaten?	positive	joy	null	casual	no	matched	Care expression very common	food	neutral	1	original	2026-04-23
WZ-T-0004	You don try	Well done — sincere	positive	joy	null	casual	no	matched	Sincere warm praise	social	neutral	1	original	2026-04-23
WZ-T-0005	You don try oh	You really did well	positive	joy	null	casual	no	matched	Oh increases sincerity and warmth	social	neutral	2	original	2026-04-23
WZ-T-0006	You don try well well	You did extremely well — sincere	positive	celebration	null	casual	no	matched	Intensifier doubles sincere praise	social	neutral	3	original	2026-04-23
WZ-T-0007	You don try well well	You think you did well? — sarcastic	negative	forming	null	street	yes	contradicts	Speaker sensed prior deceit or failure — same phrase opposite meaning	social	neutral	3	original	2026-04-23
WZ-T-0008	You don really try	You genuinely did well	positive	celebration	null	casual	no	matched	Really as sincere intensifier	social	neutral	2	original	2026-04-23
WZ-T-0009	E don do abeg	I've had enough please — exhaustion	negative	hustle_fatigue	null	casual	no	matched	Exhaustion and resignation	emotion	neutral	2	original	2026-04-23
WZ-T-0010	E don do	That's enough — stop quantity	neutral	neutral	null	casual	no	matched	Stopping someone from adding too much — food money quantity etc	life	neutral	1	original	2026-04-23
WZ-T-0011	E don do	It's okay — consoling someone crying	neutral	neutral	null	casual	no	matched	Consoling someone upset or crying — gentle and warm	emotion	neutral	1	original	2026-04-23
WZ-T-0012	E don do	Stop it — breaking up a fight	negative	anger	null	street	no	matched	Commanding someone to stop fighting or arguing	conflict	neutral	2	original	2026-04-23
WZ-T-0013	E don do	Enough explanation — cut it short	neutral	neutral	null	casual	no	matched	Cutting off excessive talking or explanation	social	neutral	1	original	2026-04-23
WZ-T-0014	Omo e hard o	Wow it's really difficult	negative	hustle_fatigue	null	casual	no	matched	Acknowledging struggle	emotion	neutral	2	original	2026-04-23
WZ-T-0015	I no fit shout	I can't be bothered	negative	hustle_fatigue	null	street	no	matched	Too tired to react	emotion	neutral	2	original	2026-04-23
WZ-T-0016	Heat dey catch me	The heat is getting to me	negative	hustle_fatigue	null	casual	no	matched	Mild heat complaint	life	neutral	1	original	2026-04-23
WZ-T-0017	Heat wan wound me	This heat is really bad	negative	hustle_fatigue	null	street	no	matched	Dramatic intensification	life	neutral	2	original	2026-04-23
WZ-T-0018	Heat wan kill me	This heat is absolutely unbearable	negative	hustle_fatigue	null	street	no	matched	Peak dramatic expression	life	neutral	3	original	2026-04-23
WZ-T-0019	Na God o	It's God's doing	positive	prayer_gratitude	null	casual	no	matched	Gratitude expression	faith	neutral	1	original	2026-04-23
WZ-T-0020	God don butter my bread	God has blessed me	positive	prayer_gratitude	null	casual	no	matched	Gratitude for breakthrough	faith	neutral	2	original	2026-04-23
WZ-T-0021	You too dey form	You're acting too cool/fake	negative	forming	null	street	no	matched	Calling out pretension	social	neutral	2	original	2026-04-23
WZ-T-0022	You well at all?	Are you okay at all?	negative	anger	null	casual	no	matched	Mild rebuke and disbelief	social	neutral	2	original	2026-04-23
WZ-T-0023	Wahala dey o	There's trouble/drama	negative	anger	null	casual	no	matched	Warning of incoming problem	social	neutral	2	original	2026-04-23
WZ-T-0024	Wetin you mean?	What do you mean?	negative	suspicion	null	casual	no	matched	Challenging someone's statement	social	neutral	2	original	2026-04-23
WZ-T-0025	This man sef	Ugh this man	negative	anger	null	casual	no	matched	Standalone exasperation — complete thought	social	neutral	2	original	2026-04-23
WZ-T-0026	This man eh	Ugh this man	negative	anger	null	casual	no	matched	Same meaning as this man sef — different expression	social	neutral	2	original	2026-04-23
WZ-T-0027	This boy dey gba	This boy is smelling	negative	anger	null	street	no	matched	Direct social judgment about body odour	social	neutral	2	original	2026-04-23
WZ-T-0028	You never lock door	You haven't locked the door	neutral	neutral	null	casual	no	matched	Plain reminder no emotional charge	life	neutral	1	original	2026-04-23
WZ-T-0029	You never lock door oh	You still haven't locked the door	neutral	neutral	null	casual	no	matched	Emotion escalates to anger depending on tone — see prosody	life	neutral	2	original	2026-04-23
WZ-T-0030	Ordinary water he no get	He doesn't even have basic water	negative	shock	null	street	no	matched	Shock at complete unpreparedness — betrayal also present	social	neutral	3	original	2026-04-23
WZ-T-0031	You just dey come Lagos?	Are you new to Lagos? — genuine	neutral	neutral	null	casual	no	matched	Genuine question to newcomer	social	neutral	1	original	2026-04-23
WZ-T-0032	You just dey come Lagos?	You clearly don't know Lagos — mockery	negative	forming	null	street	yes	contradicts	Mockery of naivety — speaker still in Lagos context	social	neutral	2	original	2026-04-23
WZ-T-0033	Abeg leave matter	Please drop the topic	negative	hustle_fatigue	null	casual	no	matched	Dismissal tired of subject	emotion	neutral	2	original	2026-04-23
WZ-T-0034	E be things	That's just how it is	neutral	neutral	null	casual	no	matched	Resignation and acceptance	life	neutral	1	original	2026-04-23
WZ-T-0035	I dey kampe	I'm doing very well	positive	joy	null	casual	no	matched	Strong positive state	greeting	neutral	1	original	2026-04-23
WZ-T-0036	Wetin dey happen?	What's going on?	neutral	neutral	null	casual	no	matched	General check-in	social	neutral	1	original	2026-04-23
WZ-T-0037	We move!	We're moving forward!	positive	celebration	null	street	no	matched	Hype and motivation phrase	motivation	neutral	3	original	2026-04-23
WZ-T-0038	Dem never see anything	They haven't seen anything yet	positive	pride	null	street	no	matched	Confidence and swagger	motivation	neutral	2	original	2026-04-23
WZ-T-0039	No dulling we move	No slacking let's go	positive	celebration	null	street	no	matched	Motivational push	motivation	neutral	3	original	2026-04-23
WZ-T-0040	E sweet me die	I enjoyed it so much	positive	joy	null	street	no	matched	Intense enjoyment	emotion	neutral	2	original	2026-04-23
WZ-T-0041	You wan go market?	Do you want to go to the market?	neutral	neutral	null	casual	no	matched	Simple question no emotional charge	life	neutral	1	original	2026-04-23
WZ-T-0042	Abi now	You agree right? — seeking confirmation	neutral	neutral	null	casual	no	matched	Seeking agreement or validation from listener	social	neutral	1	original	2026-04-23
WZ-T-0043	You this person	You this person — frustrated	negative	anger	null	casual	no	matched	Real frustration and irritation — not playful	social	neutral	2	original	2026-04-23
WZ-T-0044	E be like say	It seems like — realization	neutral	neutral	null	casual	no	matched	Realization version — speaker just figured something out	social	neutral	1	original	2026-04-23
WZ-T-0045	E be like say	It seems like — doubt	negative	suspicion	null	casual	no	matched	Doubt version — speaker not convinced	social	neutral	2	original	2026-04-23
WZ-T-0046	No be small thing	This exceeded expectations	positive	pride	null	casual	no	matched	Happy about outcome — slightly exceeded expectation or genuine admiration	social	neutral	2	original	2026-04-23
WZ-T-0047	See your life	Look at the state of you	negative	forming	null	street	no	matched	Pure mockery — no affectionate version	social	neutral	2	original	2026-04-23
WZ-T-0048	I hail	I respect you	positive	pride	null	casual	no	matched	Pure respect and acknowledgement — never sarcastic	social	neutral	1	original	2026-04-23
WZ-T-0049	Shine your eye	Be alert — warning of danger or deceit	negative	suspicion	null	street	no	matched	Warning for deceitful cunning unpredictable or risky situations	social	neutral	2	original	2026-04-23
WZ-T-0050	Shine your eye well well	Be very alert — strong warning	negative	suspicion	null	street	no	matched	Intensified version — situation is seriously risky or dangerous	social	neutral	3	original	2026-04-23
WZ-T-0051	How much be the money?	How much does it cost?	neutral	neutral	null	casual	no	matched	Simple price inquiry	money	neutral	1	original	2026-04-23
WZ-T-0052	Allawee don show	Allowance has arrived	positive	celebration	null	casual	no	matched	Joy at receiving allowance	money	neutral	2	original	2026-04-23
WZ-T-0053	Salary don show	Salary has arrived	positive	celebration	null	casual	no	matched	Joy at receiving salary	money	neutral	3	original	2026-04-23
WZ-T-0054	I no get money	I don't have money	negative	hustle_fatigue	null	casual	no	matched	Plain statement of being broke	money	neutral	1	original	2026-04-23
WZ-T-0055	Owu dey blow me	Poverty is hitting me hard	negative	hustle_fatigue	null	street	no	matched	Humorous informal word for poverty — same as sapa	money	neutral	2	original	2026-04-23
WZ-T-0056	Sapa dey	There's no money — poverty is present	negative	hustle_fatigue	null	street	no	matched	Sapa = symbolic word for poverty — dey confirms its presence	money	neutral	2	original	2026-04-23
WZ-T-0057	Sapa	Poverty — I am broke	negative	hustle_fatigue	null	street	no	matched	Standalone sapa — more intense — I AM the poverty	money	neutral	3	original	2026-04-23
WZ-T-0058	I no hold money	I don't have money	negative	hustle_fatigue	null	casual	no	matched	Money comes and goes — can't retain it	money	neutral	2	original	2026-04-23
WZ-T-0059	Comot money for me now?	Give me a discount	neutral	market_energy	null	casual	no	matched	Requesting discount — direct ask	money	neutral	1	original	2026-04-23
WZ-T-0060	No charge me oh	Don't give me an expensive price	negative	market_energy	null	casual	no	matched	Requesting not to be overcharged — mild plea	money	neutral	1	original	2026-04-23
WZ-T-0061	You too cost	You are too expensive	negative	market_energy	null	street	no	matched	Frustration at high price of person or thing	money	neutral	2	original	2026-04-23
WZ-T-0062	E cost oh	It's expensive	negative	market_energy	null	casual	no	matched	General complaint about price	money	neutral	2	original	2026-04-23
WZ-T-0063	I dey owe	I am in debt	negative	hustle_fatigue	null	casual	no	matched	Stating debt situation plainly	money	neutral	1	original	2026-04-23
WZ-T-0064	I no fit pay	I can't afford to pay	negative	hustle_fatigue	null	casual	no	matched	Inability to pay — resigned	money	neutral	2	original	2026-04-23
WZ-T-0065	Come collect your money	Come and take your money	neutral	neutral	null	casual	no	matched	Telling someone to collect what is owed to them	money	neutral	1	original	2026-04-23
WZ-T-0066	Bank debit me yesterday	The bank debited me yesterday	negative	shock	null	casual	no	matched	Surprise and frustration at unexpected debit	money	neutral	2	original	2026-04-23
WZ-T-0067	Opay dem debit me yesterday	Opay debited me yesterday	negative	shock	null	street	no	matched	Same as bank debit but specific to Opay — very natural Lagos fintech speech	money	neutral	2	original	2026-04-23
WZ-T-0068	Credit alert dey hungry me	I really want to get a credit alert	negative	craving	null	street	no	matched	Intense longing for money to land in account	money	neutral	3	original	2026-04-23
WZ-T-0069	I never get alert	I haven't received an alert yet	negative	craving	null	casual	no	matched	Waiting anxiously for payment to arrive	money	neutral	2	original	2026-04-23
WZ-T-0070	You no go comot money for me?	Won't you give me a discount?	negative	suspicion	null	casual	no	matched	Pleading with frustration — rhetorical discount request	money	neutral	2	original	2026-04-23
WZ-T-0071	Wetin you wan buy?	What do you want to buy?	neutral	neutral	null	casual	no	matched	Seller asking buyer intent	money	neutral	1	original	2026-04-23
WZ-T-0072	Wetin you dey sell?	What are you selling?	neutral	neutral	null	casual	no	matched	Buyer asking seller what they have	money	neutral	1	original	2026-04-23
WZ-T-0073	Your price fit kill persin	Your price can kill someone	negative	market_energy	null	street	no	matched	Rhetorical frustration at extremely high price	money	neutral	3	original	2026-04-23
WZ-T-0074	Persin no fit buy market from your hand oh?	Someone can't buy something from you oh	negative	market_energy	null	street	no	matched	Rhetorical frustration — prices too high for anyone to afford	money	neutral	3	original	2026-04-23
WZ-T-0075	No send me go village oh	Don't make me broke	negative	suspicion	null	street	no	matched	Warning not to drain someone financially	money	neutral	2	original	2026-04-23
WZ-T-0076	How much be d thing?	How much is it?	neutral	neutral	null	casual	no	matched	Casual price inquiry — d is the	money	neutral	1	original	2026-04-23
WZ-T-0077	No dey owe persin	Don't owe people	neutral	neutral	null	casual	no	matched	Advice against owing people — proverb-like	money	neutral	1	original	2026-04-23
WZ-T-0078	You too beg	You beg too much	negative	anger	null	street	no	matched	Frustration at someone who constantly asks for things	money	neutral	2	original	2026-04-23
WZ-T-0079	Abeg help me transfer money	Please help me transfer money	neutral	neutral	null	casual	no	matched	Polite request for transfer help	money	neutral	1	original	2026-04-23
WZ-T-0080	Transfer no dey go	The transfer is not going through	negative	anger	null	casual	no	matched	Frustration at failed transfer — very common Lagos fintech pain	money	neutral	2	original	2026-04-23
WZ-T-0081	Bank never open	The bank hasn't opened yet	neutral	neutral	null	casual	no	matched	Stating bank is not yet open	money	neutral	1	original	2026-04-23
WZ-T-0082	You don pay?	Have you paid?	neutral	neutral	null	casual	no	matched	Simple payment confirmation check	money	neutral	1	original	2026-04-23
WZ-T-0083	Card no dey work oh	My card is not working	negative	anger	null	casual	no	matched	Frustration at card failure — common fintech pain point	money	neutral	2	original	2026-04-23
WZ-T-0084	Abeg transfer my money	Please transfer my money	neutral	neutral	null	casual	no	matched	Direct request for transfer	money	neutral	1	original	2026-04-23
WZ-T-0085	I no fit buy am that price	I can't buy it at that price	negative	market_energy	null	casual	no	matched	Price negotiation — rejecting current price	money	neutral	2	original	2026-04-23
WZ-T-0086	No be small money oh	This is not a small amount of money	negative	shock	null	casual	no	matched	Expressing that the amount is significant	money	neutral	2	original	2026-04-23
WZ-T-0087	Money no dey	There is no money	negative	hustle_fatigue	null	casual	no	matched	Plain statement — broke	money	neutral	2	original	2026-04-23
WZ-T-0088	I no fit buy am	I cannot afford it	negative	hustle_fatigue	null	casual	no	matched	Simple inability to afford — resigned	money	neutral	1	original	2026-04-23
WZ-T-0089	You never pay me	You haven't paid me yet	negative	anger	null	casual	no	matched	Demanding unpaid debt — direct	money	neutral	2	original	2026-04-23
WZ-T-0090	Where my money dey?	Where is my money?	negative	anger	null	casual	no	matched	Demanding owed money — escalated	money	neutral	3	original	2026-04-23
WZ-T-0091	E don enter my account	It has entered my account	positive	celebration	null	casual	no	matched	Joy at money landing — alert received	money	neutral	3	original	2026-04-23
WZ-T-0092	Them block my account	They blocked my account	negative	shock	null	casual	no	matched	Shock at account being frozen — common fintech pain	money	neutral	3	original	2026-04-23
WZ-T-0093	I go pay you back	I will pay you back	neutral	neutral	null	casual	no	matched	Promise to repay debt	money	neutral	1	original	2026-04-23
WZ-T-0094	When you go pay me?	When are you going to pay me?	negative	anger	null	casual	no	matched	Frustrated creditor demanding repayment timeline	money	neutral	2	original	2026-04-23
WZ-T-0095	E cheap	It's cheap	positive	market_energy	null	casual	no	matched	Positive reaction to affordable price	money	neutral	1	original	2026-04-23
WZ-T-0096	Make we price am	Let's negotiate the price	neutral	market_energy	null	casual	no	matched	Opening price negotiation — very Lagos market	money	neutral	1	original	2026-04-23
WZ-T-0097	Last price	What's your final price?	neutral	market_energy	null	street	no	matched	Final negotiation push — buyer demanding best price	money	neutral	2	original	2026-04-23
WZ-T-0098	I go manage am	I will make do with it	negative	hustle_fatigue	null	casual	no	matched	Accepting something despite it not being ideal	money	neutral	1	original	2026-04-23
WZ-T-0099	No be here I go find money	This is not where I will make money	negative	pride	null	street	no	matched	Rejecting a bad deal with confidence — walking away energy	money	neutral	2	original	2026-04-23
WZ-T-0100	My account don red	My account is overdrawn	negative	shock	null	casual	no	matched	Account overdrawn or empty — visual metaphor	money	neutral	3	original	2026-04-23

End of preview.

Wazobia Labs — Nigerian Pidgin (Naija) Sentiment & Emotion Dataset

Version: v0.6 | Entries: 484 | License: CC-BY-4.0 | Builder: Wazobia Labs

Using this dataset in research or a product? Please reach out: wazobialabs@gmail.com

Dataset Summary

This dataset is the foundational corpus for Nigerian Pidgin (Naija) sentiment and emotion classification — built to fill the specific gaps that every other African language dataset leaves open.

Most available Nigerian language datasets are built from formal sources: news broadcasts, scripted studio recordings, parliamentary transcripts. This dataset captures a different register entirely — the language Nigerians actually use on WhatsApp, at the market, at home, during prayer, in conflict, and in joy.

This is not another Twitter scrape. Not another BBC corpus. Not another dataset that expires on import.

This dataset is the text foundation for an upcoming gold-standard Nigerian Pidgin Evaluation Set (target Q3 2026) — benchmark infrastructure for any AI lab building Nigerian language products.

What Makes This Dataset Different

Gap in Existing Data	What This Dataset Builds
ASR transcription only — no emotional layer	Sentiment + prosody match annotation. Does tone match meaning? Sarcasm flagged explicitly.
Scripted studio recordings only	Natural WhatsApp-register speech — spontaneous, unscripted
Code-switching treated as noise	Register annotated: casual, street, proverbial
Positive / negative / neutral only	16-category Nigerian emotion taxonomy including forming, hustle_fatigue, hustle_energy, market_energy, prayer_gratitude, sarcasm, contempt
Words transcribed, pitch ignored	Prosody match field — does voice tone match text meaning?
Datasets skew male	45% female speaker representation — hard architectural constraint
No evaluation infrastructure	39 complete sarcasm pairs with sincere twins — eval-set ready
Same phrase assumed one meaning	Tonal disambiguation entries: same phrase, multiple contexts, different emotions

Dataset Statistics — v0.6

Field	Value
Total entries	484
Version	0.6.0
Date updated	May 2026
Language	Nigerian Pidgin English (Naija / pcm)
License	CC-BY-4.0 — commercial use permitted

Speaker Gender

Gender	Count	%
Female	220	45%
Neutral / unspecified	264	55%

Topic Domains

Domain	Count
Social	98
Health	74
Money	72
Work	54
Life	35
Relationship	34
Faith	30
Conflict	25
Emotion	21
Food	20
Motivation	12
Greeting	9

Emotion Categories — 16 Total

Emotion	Count	Notes
neutral	87	No specific emotion detectable
hustle_fatigue	61	Exhaustion from grinding, struggling, enduring
pride	45	Confidence, swagger, self-assurance
forming	42	Performing coolness — acting unbothered on purpose
anger	39	Frustration, irritation, outrage
prayer_gratitude	34	Spiritual thankfulness, testimony, acknowledging God's hand
joy	32	Genuine happiness, excitement, delight
celebration	25	Hype, wins, breakthroughs, collective joy
betrayal	21	Feeling deceived or disrespected by someone trusted
suspicion	20	Awareness of attempted manipulation or deception
shock	20	Surprise mixed with disbelief
market_energy	20	Sharp transactional Lagos commerce energy
craving	20	Intense desire — money, success, validation, breakthrough
sarcasm	8	Saying opposite of what you mean — v0.6 introduction
contempt	7	Cold downward dismissal — v0.6 introduction
hustle_energy	3	Motivated grind energy — v0.6 introduction

Note: sarcasm, contempt, and hustle_energy are new categories introduced in v0.6. All 39 sarcasm pairs have sarcasm_flag=yes regardless of emotion_category.

Sarcasm

Flag	Count
Sarcasm pairs (yes)	39
Sincere entries (no)	445

Every sarcasm entry has a sincere twin — an identical phrase annotated with its genuine non-sarcastic meaning. This paired structure is the foundation of the evaluation set.

Register

Register	Count
Casual	355
Street	122
Proverbial	7

Sentiment

Sentiment	Count
Negative	259
Positive	150
Neutral	75

Nigerian Emotion Taxonomy — 16 Categories

This taxonomy is Wazobia Labs' most defensible intellectual asset. It cannot be built without lived Nigerian cultural experience.

Original 13 Categories

Category	Definition
joy	Genuine happiness, excitement, delight
anger	Frustration, irritation, outrage
forming	Performing coolness — acting unbothered on purpose
hustle_fatigue	Exhaustion from grinding, struggling, enduring — e don do energy
market_energy	Sharp transactional Lagos commerce energy
prayer_gratitude	Spiritual thankfulness, testimony, acknowledging God's hand
betrayal	Feeling deceived or disrespected by someone trusted
celebration	Hype, wins, breakthroughs, collective joy
pride	Confidence, swagger, self-assurance
suspicion	Awareness of attempted manipulation or deception
shock	Surprise mixed with disbelief — how can this be real energy
craving	Intense desire or hunger — money, success, validation, breakthrough
neutral	No specific emotion detectable

New in v0.6

Category	Definition
sarcasm	Saying the opposite of what you mean to wound, dismiss, or mock — delivery contradicts content
contempt	Cold downward dismissal — not angry, not performing, just done. Looking down not across.
hustle_energy	Motivated grind energy — fired up, eyes on the prize, nothing go stop me

Key distinctions:

forming vs contempt: forming is performed indifference. Contempt is genuine cold dismissal.
hustle_fatigue vs hustle_energy: same grind context, opposite emotional state.
sarcasm vs forming: forming performs a mood. Sarcasm inverts meaning through delivery.

Data Schema

Column	Type	Description
`entry_id`	string	Unique ID — format WZ-T-XXXX
`pidgin_text`	string	Original Nigerian Pidgin utterance
`english_gloss`	string	English translation / gloss
`sentiment`	string	positive / negative / neutral
`emotion_category`	string	Primary emotion from 16-category taxonomy
`emotion_secondary`	string	Secondary emotion where applicable
`register`	string	casual / street / proverbial
`sarcasm_flag`	string	yes / no
`prosody_match`	string	matched / contradicts
`annotator_notes`	string	Cultural and linguistic context notes
`topic_domain`	string	social / health / money / work / relationship / faith / life / conflict / food / emotion / motivation / greeting
`speaker_gender`	string	female / neutral
`intensity`	integer	1 (low) / 2 (medium) / 3 (high)
`source_type`	string	original
`date_added`	date	YYYY-MM-DD

Sarcasm Pairs — Evaluation Infrastructure

A defining feature of this dataset is 39 complete sarcasm pairs, each consisting of an identical Pidgin phrase annotated once as sincere and once as sarcastic.

Example:

entry_id	pidgin_text	sarcasm_flag	prosody_match	emotion_category
WZ-T-0006	You don try well well	no	matched	celebration
WZ-T-0007	You don try well well	yes	contradicts	forming

The same words carry opposite meanings depending on delivery. This is the core challenge for AI systems processing Nigerian Pidgin.

Tonal Disambiguation Entries

entry_id	pidgin_text	context	emotion
WZ-T-0010	E don do	Stopping excess — food / money	neutral
WZ-T-0011	E don do	Consoling someone crying	neutral
WZ-T-0012	E don do	Commanding stop during a fight	anger
WZ-T-0013	E don do	Cutting off excessive talking	neutral

These are intentional contextual variants demonstrating Gap 5 — tonal disambiguation. Not duplicates.

Intended Uses

Recommended:

Sentiment and emotion classification for Nigerian Pidgin
Sarcasm detection benchmarking
Fine-tuning multilingual models (mBERT, XLM-R, AfroXLMR) on Nigerian Pidgin
Evaluating AI models for Nigerian language products
Low-resource NLP research on West African languages
Cultural nuance and register analysis

Not recommended without domain adaptation:

Formal or literary Pidgin processing
Tasks requiring perfectly balanced class distributions

Version Notes — v0.6

Voice data releases in v1.0 (Q4 2026). Female speaker representation is at 45% and actively growing toward the 50% hard target. Inter-annotator review with a second native speaker is in progress ahead of v1.0. The three new emotion categories — sarcasm, contempt, and hustle_energy — were introduced this version and will reach full coverage in v0.7. Full sarcasm infrastructure is captured in the sarcasm_flag=yes column (39 complete pairs).

Roadmap

Version	Target	Key Additions
v0.7	July 2026	50% female, Igbo dataset launched, new categories expanded
v1.0	Q4 2026	2,000+ entries, 1,000 voice recordings, prosody annotations, inter-annotator scores
Eval Set	Q3 2026	Separate repo — 200 gold-standard entries, min 20 per original 13 emotion categories

Citation

@dataset{wazobia_labs_pidgin_2026,
  author    = {Okoye, Stephanie},
  title     = {Wazobia Labs Nigerian Pidgin Sentiment and Emotion Dataset},
  year      = {2026},
  version   = {0.6.0},
  publisher = {Hugging Face},
  url       = {https://huggingface.co/datasets/WAZOBIALABS/nigerian-pidgin-voice-text},
  license   = {CC-BY-4.0}
}

Using This Dataset?

If you are using this dataset in research or building a product with it, please reach out — wazobialabs@gmail.com

We are actively building the Nigerian Pidgin Evaluation Set and want to know what the community is building.

About Wazobia Labs

Wazobia Labs builds African language AI infrastructure that doesn't exist but should. We identify the specific, high-value gaps in African language data that existing datasets leave open — then build exactly those gaps with commercial licensing, production-grade quality, and the cultural specificity that real AI products need.

Contact: wazobialabs@gmail.com Hugging Face: WAZOBIALABS Founded: Lagos, Nigeria — 2025

Wa. Zo. Bia. The data is ready.

Downloads last month: 223

Total file size:

220 kB