Dataset Preview
The full dataset viewer is not available (click to read why). Only showing a preview of the rows.
The dataset generation failed
Error code: DatasetGenerationError
Exception: CastError
Message: Couldn't cast
units: list<item: string>
child 0, item: string
unit_to_class: struct<<sil>: int64, a: int64, b: int64, c: int64, d: int64, dw: int64, e: int64, f: int64, g: int64 (... 291 chars omitted)
child 0, <sil>: int64
child 1, a: int64
child 2, b: int64
child 3, c: int64
child 4, d: int64
child 5, dw: int64
child 6, e: int64
child 7, f: int64
child 8, g: int64
child 9, gy: int64
child 10, h: int64
child 11, hw: int64
child 12, hy: int64
child 13, i: int64
child 14, j: int64
child 15, k: int64
child 16, kw: int64
child 17, ky: int64
child 18, l: int64
child 19, m: int64
child 20, n: int64
child 21, nw: int64
child 22, ny: int64
child 23, o: int64
child 24, p: int64
child 25, r: int64
child 26, s: int64
child 27, t: int64
child 28, tw: int64
child 29, u: int64
child 30, v: int64
child 31, w: int64
child 32, x: int64
child 33, y: int64
child 34, z: int64
child 35, ɔ: int64
child 36, ɛ: int64
frame_counts: struct<a: int64, n: int64, e: int64, ɛ: int64, o: int64, s: int64, i: int64, m: int64, t: int64, b: (... 298 chars omitted)
child 0, a: int64
child 1, n: int64
child 2, e: int64
child 3, ɛ: int64
child 4, o: int64
child 5, s: int64
child 6, i: int64
child 7, m: int64
child 8, t: int64
child 9, b: int64
child 10, ɔ: int64
child 11, k: int64
child 12, d: int64
child 13, u: int64
child 14, r: int64
child 15, y: int64
child 16, f: int64
child 17, w: int64
child 18, p: int64
c
...
int64, o: int64, m: int64, s: int64, i: int64, ɔ: int64, b (... 298 chars omitted)
child 0, a: int64
child 1, n: int64
child 2, ɛ: int64
child 3, e: int64
child 4, o: int64
child 5, m: int64
child 6, s: int64
child 7, i: int64
child 8, ɔ: int64
child 9, b: int64
child 10, t: int64
child 11, d: int64
child 12, k: int64
child 13, y: int64
child 14, u: int64
child 15, w: int64
child 16, r: int64
child 17, f: int64
child 18, p: int64
child 19, h: int64
child 20, l: int64
child 21, ny: int64
child 22, g: int64
child 23, c: int64
child 24, ky: int64
child 25, v: int64
child 26, hw: int64
child 27, kw: int64
child 28, gy: int64
child 29, dw: int64
child 30, hy: int64
child 31, tw: int64
child 32, j: int64
child 33, nw: int64
child 34, z: int64
child 35, x: int64
child 36, q: int64
child 37, gw: int64
config: struct<chunk_ms: int64, chunk_stride_ms: int64, min_unit_score: double, min_utt_score: double, min_u (... 161 chars omitted)
child 0, chunk_ms: int64
child 1, chunk_stride_ms: int64
child 2, min_unit_score: double
child 3, min_utt_score: double
child 4, min_unit_count: int64
child 5, silence_as_ignore: bool
child 6, silence_energy_percentile: double
child 7, min_silence_gap_ms: int64
child 8, max_chunks_per_split: int64
child 9, splits: list<item: string>
child 0, item: string
mean_score: double
n_frames: int64
text: string
n_units: int64
utt: string
row: int64
duration: double
to
{'utt': Value('string'), 'row': Value('int64'), 'duration': Value('float64'), 'n_frames': Value('int64'), 'n_units': Value('int64'), 'mean_score': Value('float64'), 'text': Value('string')}
because column names don't match
Traceback: Traceback (most recent call last):
File "/usr/local/lib/python3.14/site-packages/datasets/builder.py", line 1827, in _prepare_split_single
for key, table in generator:
^^^^^^^^^
File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 613, in wrapped
for item in generator(*args, **kwargs):
~~~~~~~~~^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/json/json.py", line 343, in _generate_tables
self._cast_table(pa_table, json_field_paths=json_field_paths),
~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/json/json.py", line 132, in _cast_table
pa_table = table_cast(pa_table, self.info.features.arrow_schema)
File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2378, in table_cast
return cast_table_to_schema(table, schema)
File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2306, in cast_table_to_schema
raise CastError(
...<3 lines>...
)
datasets.table.CastError: Couldn't cast
units: list<item: string>
child 0, item: string
unit_to_class: struct<<sil>: int64, a: int64, b: int64, c: int64, d: int64, dw: int64, e: int64, f: int64, g: int64 (... 291 chars omitted)
child 0, <sil>: int64
child 1, a: int64
child 2, b: int64
child 3, c: int64
child 4, d: int64
child 5, dw: int64
child 6, e: int64
child 7, f: int64
child 8, g: int64
child 9, gy: int64
child 10, h: int64
child 11, hw: int64
child 12, hy: int64
child 13, i: int64
child 14, j: int64
child 15, k: int64
child 16, kw: int64
child 17, ky: int64
child 18, l: int64
child 19, m: int64
child 20, n: int64
child 21, nw: int64
child 22, ny: int64
child 23, o: int64
child 24, p: int64
child 25, r: int64
child 26, s: int64
child 27, t: int64
child 28, tw: int64
child 29, u: int64
child 30, v: int64
child 31, w: int64
child 32, x: int64
child 33, y: int64
child 34, z: int64
child 35, ɔ: int64
child 36, ɛ: int64
frame_counts: struct<a: int64, n: int64, e: int64, ɛ: int64, o: int64, s: int64, i: int64, m: int64, t: int64, b: (... 298 chars omitted)
child 0, a: int64
child 1, n: int64
child 2, e: int64
child 3, ɛ: int64
child 4, o: int64
child 5, s: int64
child 6, i: int64
child 7, m: int64
child 8, t: int64
child 9, b: int64
child 10, ɔ: int64
child 11, k: int64
child 12, d: int64
child 13, u: int64
child 14, r: int64
child 15, y: int64
child 16, f: int64
child 17, w: int64
child 18, p: int64
c
...
int64, o: int64, m: int64, s: int64, i: int64, ɔ: int64, b (... 298 chars omitted)
child 0, a: int64
child 1, n: int64
child 2, ɛ: int64
child 3, e: int64
child 4, o: int64
child 5, m: int64
child 6, s: int64
child 7, i: int64
child 8, ɔ: int64
child 9, b: int64
child 10, t: int64
child 11, d: int64
child 12, k: int64
child 13, y: int64
child 14, u: int64
child 15, w: int64
child 16, r: int64
child 17, f: int64
child 18, p: int64
child 19, h: int64
child 20, l: int64
child 21, ny: int64
child 22, g: int64
child 23, c: int64
child 24, ky: int64
child 25, v: int64
child 26, hw: int64
child 27, kw: int64
child 28, gy: int64
child 29, dw: int64
child 30, hy: int64
child 31, tw: int64
child 32, j: int64
child 33, nw: int64
child 34, z: int64
child 35, x: int64
child 36, q: int64
child 37, gw: int64
config: struct<chunk_ms: int64, chunk_stride_ms: int64, min_unit_score: double, min_utt_score: double, min_u (... 161 chars omitted)
child 0, chunk_ms: int64
child 1, chunk_stride_ms: int64
child 2, min_unit_score: double
child 3, min_utt_score: double
child 4, min_unit_count: int64
child 5, silence_as_ignore: bool
child 6, silence_energy_percentile: double
child 7, min_silence_gap_ms: int64
child 8, max_chunks_per_split: int64
child 9, splits: list<item: string>
child 0, item: string
mean_score: double
n_frames: int64
text: string
n_units: int64
utt: string
row: int64
duration: double
to
{'utt': Value('string'), 'row': Value('int64'), 'duration': Value('float64'), 'n_frames': Value('int64'), 'n_units': Value('int64'), 'mean_score': Value('float64'), 'text': Value('string')}
because column names don't match
The above exception was the direct cause of the following exception:
Traceback (most recent call last):
File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 1369, in compute_config_parquet_and_info_response
parquet_operations, partial, estimated_dataset_info = stream_convert_to_parquet(
~~~~~~~~~~~~~~~~~~~~~~~~~^
builder, max_dataset_size_bytes=max_dataset_size_bytes
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
)
^
File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 948, in stream_convert_to_parquet
builder._prepare_split(split_generator=splits_generators[split], file_format="parquet")
~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/builder.py", line 1694, in _prepare_split
for job_id, done, content in self._prepare_split_single(
~~~~~~~~~~~~~~~~~~~~~~~~~~^
gen_kwargs=gen_kwargs, job_id=job_id, **_prepare_split_args
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
):
^
File "/usr/local/lib/python3.14/site-packages/datasets/builder.py", line 1880, in _prepare_split_single
raise DatasetGenerationError("An error occurred while generating the dataset") from e
datasets.exceptions.DatasetGenerationError: An error occurred while generating the datasetNeed help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.
utt string | row int64 | duration float64 | n_frames int64 | n_units int64 | mean_score float64 | text string |
|---|---|---|---|---|---|---|
agric_00000_000000 | 0 | 30 | 3,001 | 202 | -0.5857 | na ɛsɛ transport officer okay na na adwuma no a na mereyɛ no na kwaadwuma no nso mede abata ho na adwuma no broni bɛkaa sɛ na ɛyɛ mede gyegye m'ani oh okay a yɛbɛfrɛ no sɛ hall bi nti nyɛ adeɛ sɛ mekɔ agreed school bi ɛkɔ suae anaasɛ merekɔsua no deɛ na ɔmo nyinaa yɛ |
agric_00000_000001 | 1 | 30 | 3,001 | 187 | -0.8273 | ɔhyeɛfoɔ ɛyɛ cocoa a ɛno aba abɛyɛ nsɛm nna hɔ adwuma a woyɛeɛ woakyerɛkyerɛ sɛ yɛduru baabi koraa na wo de cup ne ade na ɛyɛ irrigation ɛno yɛde adikan deɛ ɔde yɛpɛ hu no sɛ woyɛ adwuma na adɔ ne ne nkeka ho ne adeɛ no na ɛte sɛn because bi da |
agric_00000_000002 | 2 | 30 | 3,001 | 166 | -1.0417 | ɔmo bɛ twaa ɔmo bɛma yɛ cocoa nurse na yɛde adua na ɔmo ama yɛn borɔdeɛ nti ɔmo bɛyɛ lie and pergen line mpɛ no bɛyɛ 10 feet 10 feet no yɛ wie a na yɛama wo cocoa na ɔde asi borɔdeɛ no nkyɛn enti mmom deɛ manya nto because |
agric_00000_000003 | 3 | 30 | 3,001 | 171 | -0.7599 | ɛbɛ hyɛ ɔmo nkuran esiane sɛ pollination aba abɛsu ghana sin paa ama yɛ cocoa ɛrekɔso ayekoo amen agya mema ei mepawokyɛw wo din ɛne adwuma a ɔyɛ yoo yɛfrɛ me clemente yammoa meyɛ cocoa adwuma na mesan ayɛ account hɔ |
agric_00000_000004 | 4 | 30 | 3,001 | 183 | -0.9316 | dɛn one for life na adɛn na heart ɛna mofaeɛ yes ɛɛ one for life no deɛ ɛyɛ akoma because yɛhwehwɛɛ mu ara ɛyɛ me sɛ ɔpɛ mpaninfoɔ company no bɔɔl no ɛtenase ɛhefa ɛna ɛba societal leader yɛbɛtumi complain sɛ daabi akoma yɛyɛ no sɛ |
agric_00000_000005 | 5 | 30 | 3,001 | 190 | -0.5556 | tete wo tia meyɛ plan breeder na ɛberɛ biara no yɛ adwuma no deɛ yɛyɛ no sɛ yɛ ne akuafoɔ ɛne ɔmo a ɔmo de nkate a ɛyɛ ɔmo adwuma ebia ɔmo a ɔmo tɔn ne ɔmo a ɔmo de ɛyɛ nnuane ahodoɔ ɛdi nsa wɔ soɔ ɛbisa yɛ creative variation bebree na y'aseleksi |
agric_00000_000007 | 7 | 30 | 3,001 | 172 | -0.5735 | wobɛtumi anya bi atɔ sɛdeɛn ka kyerɛ wo no sɛ interview wei akɔ akɔyɛ afebɔɔ kakra esiane sɛ 24 ɛkɔ n'awieeɛ no na utv a ɛkɔ dwuma deɛ ɛyɛ interview nti wo wura 2021 season no mu na wopɛ seedni bi adua still wokɔyɛ youtube page |
agric_00000_000008 | 8 | 30 | 3,001 | 222 | -0.7489 | yi na afei na mepɛ agyapadeɛ mennim me papa okay me papa na ɔwɔ asante region kɔkɔɔ okay enti owu wɔ hɔ no ɛna ɔmo sɛ adeɛ a costri mu yɛwoo me hɔ na yɛde me kɔɔ siri mu mekɔtena hɔ ara na brɛboɔ no nkɔ yie me ne me mmɔre no yɛ brɛ ɛna mesan ba asanteni afei na wanya mfeɛ sɛn because deɛ |
agric_00000_000009 | 9 | 30 | 3,001 | 153 | -0.5506 | woba nso a wonya bi tɔ hwɛ wohwɛfoɔ yi afoforɔ na ɛda ha yi dada yi da hɔ na wafuo bɛyɛ 800 mmɔdenbɔ sɛbe abansoro yase ɛyɛ yɛn ara nti menfa saa akwanya nso akyerɛ me nuanom a ɛhɔ ara ama no ne no a obi wɔ hɔ |
agric_00000_000011 | 11 | 30 | 3,001 | 172 | -0.79 | sɛ anka ɛyɛ wasoma bɛka akyerɛ ghanafoɔ titiriw ne mmabunu no ɔyɛ adwuma ho yoo deɛ mɛka akyerɛ titiriw ɛwɔ mmabunu ho sɛ kwa adwuma na yɛmfa no sɛ tete hɔ no sɛbi yɛse yɛkɔ school no yi yɛn time no deɛ nanka woyɛ bɔne ama |
agric_00000_000012 | 12 | 30 | 3,001 | 184 | -0.4031 | na mehunu sɛ na ɛwɔ fam na ɛwɔ fam paa enti yɛnhwɛ afei yɛmmra na twa so yɛtwa awie ɛɛ yɛ sɛdeɛ bɔhyɛ no si teɛ sɛ yɛbɛ dua bɔɔdeɛ na yɛdua cocoa ne adeɛ no yɛwieeɛ no ɛwɔ mu sɛ aban no so ɔdɔ aseɛ afa apaafoɔ adwuma wafa apaafoɔ ama ɔmo adua |
agric_00000_000013 | 13 | 30 | 3,001 | 190 | -0.6585 | kokoo no mu yie ɛba saa nso a ano no ɛha wo bebree ɛba saafoɔ yi afɔrtilize akwadaa na wobɛbɔ a esiane sɛ ahaban no yɛ bebree nti no wontumi nnya nsunsuansoɔ papa wɔ mu enti pruni no deɛ ɛyɛ adeɛ bi akuani biara afe biara at least yɛnhwɛ sɛ na |
agric_00000_000014 | 14 | 30 | 3,001 | 117 | -0.6435 | nnua aboa na nsuo atumi atɔ ama yɛn atwetwe nkɔmɔ no nyinaa yɛ kuadwuma no bi yɛn ho mmra eeiii abusuafoɔ mo akwaaba ɛɛ yɛne panin no ɛbɛtwetwe nkɔmɔ agya |
agric_00000_000016 | 16 | 30 | 3,001 | 223 | -0.7681 | mo nom nsuo paa nti no ɔnni nyinaa nsuo ɛyɛ nsuo ɛyɛ nsuo nti no wobɛhunu sɛ na emofo rolling so no yɛbisa sɛ ɔmo di afra ɛbaa ne barima na yɛhu ɔmo wɔ ha yi anaasɛ ɛyɛ mmaa nkoaa mmarima nkoa yɛka sɛ broiler a ɛyɛ brilaa okay yɛnhwɛ ɔmo abɛduru ha yi but yɛ hu sɛ ayɛ sɛ wohwɛ nkokɔ nkae |
agric_00000_000017 | 17 | 30 | 3,001 | 273 | -0.7802 | this year nana yɛpɛ sɛ wei aban ɛne development part no obiara hunu sɛ wei women with abilities no bebree naa ɛboa to provide nutrition mo kɔeɛ no deɛ bɛn ne deɛbɛyɛ no deɛ bɛn na ɔmo yɛ western region nim sɛ cocoa adwuma ahodoɔ bɛn so na yɛnuanom mmaa yi ɛreyɛ a sɛbe ɛbi mo adi dɛm so ɔmo de aduane bɛsi yɛ tables yoo ... |
agric_00000_000018 | 18 | 30 | 3,001 | 252 | -0.6814 | mu nsuo no yɛpɛ sɛ ɛba fɔm kakra na sɛneɛ mediikan kyerɛɛ mu no ɛwɔ 25 fɛnt ɛmu ɔha mu aduonu nnum nsuo a dodoɔ a ɛwɔ mu no no yɛwɔ sɛ yɛtumi ɛde ba 18 percent a ɛyɛ 18 ɔmo 18 okay nti na kyerɛ sɛ abusuafoɔ ɔkyire mu a na kyerɛ sɛ yɛbɛtoa so ahata kakra because ɔpayɛ sɛ ɛwɔ sɛ nsuo no firi mu kakra nsuo no a ɛwɔ aburo ... |
agric_00000_000019 | 19 | 30 | 3,001 | 205 | -0.5477 | adwuma koraa ɛnyɛ den biara me dɔ koraa ɛnyɛ den ama no nti obi a wobɛtumi ayɛ adi no anyɛ den wei a medɔ no mewɔ obi a ɔmo boa me nso me mma sika kakra na ɔmo aboa me afuo no nafuo no ayɛ kama wei deɛ yɛtwetwe nkɔmmɔ yi ɛyɛ den na mmaa no bi nso wɔ ha akua adwuma no ɛde ma wo |
agric_00000_000020 | 20 | 30 | 3,001 | 177 | -0.8724 | ama aburo no one ɛsa so tee ɛwɔ lines lines mu okay ɛna yɛ set plant no nso sɛ ɛbɛyɛ a ɛɛ between one plant and another no space no ɛyɛ the same ooh okay alright nti ɛboa ɛboa paa ama ne dua no mu no na yɛnhwɛ sɛ wose ɛsa so na ɔnhwɛ |
agric_00000_000021 | 21 | 30 | 3,001 | 169 | -0.6719 | ne story ne nyinaa so no yɛbɛ pɛ sɛ yɛbɛ take control no sɛn na yɛtumi control mprako no nipa dodoɔ sɛn na wo ne ɔmo yɛ adwuma yi wo work force ɛte sɛn because yɛhunu sɛ wonyin deɛ ayi kɔ dwumadie ebia daakye aba ha biom |
agric_00000_000022 | 22 | 30 | 3,001 | 189 | -0.7428 | bisa wɔ mo nea enti a ɔmo ntumi nyɛ afuo no ɛwɔ large scale like ɔmo nyɛ no kɛseɛ kɛseɛ head because obi wɔ hɔ a ɔmo yɛ no kakraa bi ɛna titiriw mmaa no nsoso no bebree no ara ntumi nyɛ like ɔmo yɛ afuo no a ɔmo yɛ no kakra nketewa monnyɛ nkɔ aries |
agric_00000_000023 | 23 | 30 | 3,001 | 185 | -0.5821 | but obi nso rehwɛ ɛbɛyɛ obi fɛ nneɛma foforɔ bɛn na 2012 yɛnhwɛ anim sɛ ministry of fishries ɛna kɔ court a mede bɛwura kuayɛ mu ɛyɛ nyame ne pɛ a 2020 yɛde nnoɔma bebree ɛtoɔ akwan mu wobɛhwɛ a other countries no ɔmo yɛ ama ɛyɛ fishes no akwannya na |
agric_00000_000024 | 24 | 30 | 3,001 | 175 | -0.3744 | erm medaase menua ɛyɛ me sɛ 2017 ansa na mebɛkɔ awors no na ayɛ gu so ara na ɛsɛ medinkan kaa no wɔ me nkasa yɛ hu sɛ moayɛ no ɛbi yɛ nantwie na ebi yɛ poultry nkokɔ adwan ne mponkye so ɛna fish pound nso ɛka ho nti mmoa yɛ nono ɛne no |
agric_00000_000025 | 25 | 30 | 3,001 | 206 | -1.0828 | ɛyɛ by day anaasɛ ɔmo te na wone ɔmo akyɛ ɔmo te na yɛakyɛ mokyerɛ mo sɛn yɛ kyɛmu 4:33 na ɛno ne nea wobɛtumi ɔmo ka no the one na yɛhwɛ a ɛnom ɔyɛ sei deɛ mebɛtua ɔmo kwa ɛno yɛ kyɛn sɛdeɛ wobɛtu ɔmo ka na asaase no ɛyɛ wo ara wo deɛ anaa wo hann me a me dea wɔato ne hann wonnhan |
agric_00000_000026 | 26 | 30 | 3,001 | 147 | -1.6405 | nananom akuafo ee akuafoɔ yɛ da mo ase o akuafoɔ yɛda mo ase naa na akuafo ɛyɛ akuafoɔ yɛ da mo ase nyansa baako ano a akuafoɔ ghana akuafo emu ayɛ bi oo akuafoɔ emu ayɛ biara eei hwan bio |
agric_00000_000028 | 28 | 30 | 3,001 | 189 | -0.6151 | kwan a ɛbɛfa so a yɛbɛtumi atete the sizes of the greenhouse no so ama ankorankoro a ɔmo yɛ ɔmo adwuma obi wɔ hɔ a ɔyɛ lawyer obi yɛ judge obi yɛ minister obi yɛ mp obi yɛ generalist nanso n ani gyei adwuma ho a ɔpɛ sɛ nka ɔno ara tɔ so dea betumi |
agric_00000_000029 | 29 | 30 | 3,001 | 198 | -0.5757 | bɛn bere bɛn kwan bɛn ɛna yɛfa so ko saa adeɛ na me mangofoɔ na mayɛ no mensuro nkɔ nkɔtwa ngu na mmom mɛtumi atoa so ayɛ okay medaase ɛnti first one no a yɛbɛ tackle no yɛ ɛɛm insects no ɛyɛ store no na ɛyɛ very important ɛnti first refu no time no yɛ sɛ |
agric_00000_000030 | 30 | 30 | 3,001 | 195 | -0.716 | monyɛ sɛn wohwɛ a wokeka bom a sɛn na monya fie na obi wɔ hɔ no ɔmo te nkran asaase yi so no aburokkeka w'ani abetɔ sɛn na first akwan no deɛ nanso afuo no nso sei me yɛ nyaa bɛyɛ 20 ɛwɔ one place pɛ mhm difference nti mhmm 21 no anaa sɛ 50 kilos de akɛseɛ no ma five |
agric_00000_000031 | 31 | 30 | 3,001 | 108 | -0.89 | region de bɛn na ɛwɔ wo tirim wokɔ w'anim da bɛn biom na wopɛ me me din na mo ima ha na woyɛ nyiniel government ayi me tɔ me kaa woamber na wa |
agric_00000_000032 | 32 | 30 | 3,001 | 198 | -0.4847 | boa ma ɔman no nyinaa atu mpɔn nti no meyɛɛ m'adwene de ɔpɛ yɛ sɛ ɛneɛ mɛba nti no anuanom na ɛnyɛ easy me first time wɔ kuro wei mu no na ɛyɛ den kakra because na asetena mu ne ade ade nti first time a mebɛduruu kuro no mu na mese ɛnneɛ mɛtumi me lie onnipa |
agric_00000_000033 | 33 | 30 | 3,001 | 163 | -0.6852 | but wonya abotare a wotumi yɛ ma no yie me ne wo yere no koraa koraa no hwɛ camera mu no aka biribi kyerɛ utv ahwɛfoɔ na dodoɔ no na titire ne ghana a mɛbu no adwuma ɛnyɛ adwuma bɔne de bɛn na ɛbɛba wiase kama yɛn adu |
agric_00000_000034 | 34 | 30 | 3,001 | 215 | -0.5042 | ɛne ɔmo a ɔmo boa ɛyɛ kua adwuma nyinaa ama ɔmo aba ama yɛabɛkyerɛkyerɛ ɔmo enti deɛ ɛdi kan a ɛno nti yɛayɛ saa modern farm wei ɛyɛ demonstration yi yɛreyɛ ama ɛyɛ akuafoɔ aba five ɛyɛ varieties akuafoɔ aba na ɔmo abɛ yi deɛ ɛbɛtumi aboa wo paa ebi wɔ hɔ nom a 85 days abɔnten |
agric_00000_000035 | 35 | 30 | 3,001 | 178 | -0.3503 | ne nkɔsuaa yɛredi ani nyinaa no time biara no ɛsɛ sɛ yɛma ɔman no ɛnam a ahoɔden wɔ mu kosuaa ahoɔden mu ebia na obi te baabi na ɔnka nsoso no ho ɛna ɔnoa nso so ɔhyɛ hɔ ɛna ɔreyɛ deɛ ɔpɛ ɛba no saa na ɛyɛ effective na anka obiara |
agric_00000_000036 | 36 | 30 | 3,001 | 276 | -0.618 | no na mmoa bi yɛfrɛ ɔmo nkakapɛ saa na yɛfrɛ ɔmo twi mu okay ɔmo ba abɛ ba borɔdeɛ no ase enti borɔdeɛ no si hɔ a wobɛhunu sɛ ewu enti ɛwɔ sɛ wopɛ sɛ saa mmoa no ɛkyɛ ansa na ɔmo aba sɛ yɛdane sebɔ a ɔmo bɛba but no si asaase no so kyɛ a ɔmo bɛba aseɛ but wopɛ sɛ ɛkyɛ ansa no ɔmo aba bɛha borɔdeɛ no deɛ a gye sɛ wotete... |
agric_00000_000037 | 37 | 30 | 3,001 | 266 | -0.4649 | fertilizer yɛde ma akuafoɔ no de bɛsii computerfoɔ no so ɛna wɔn akyɛeɛ enti okuani biara bɛdi adanseɛ sɛ fɛ laza ne nsa kaaeɛ no ɔde agu no afuo no yɛtɔ so mmienu yɛkɔfaa fidii a ɛde twa mfuo mu na wɔboroni slashers yɛde brɛ corporate so no ɔno de atwamfoɔ no mu ama ayɛ kamakama ɛba no saa a yareɛ so te ɛyɛ anonom no ... |
agric_00000_000038 | 38 | 30 | 3,001 | 184 | -0.7191 | yɛntwa yɛho mmra yɛba a yɛde nka w'atwetwe no ɛbɛhyehyɛ so ayekoo yareɛ abusua mo akwaaba sɛdeɛ mesi bɔ wo amaneɛ no ma me mma yi kuani no ɛ na afei yɛngye nkɔmɔ twetwe tie bi na mfatoho deɛ ɛboa nipa abrabɔ paa nti anka me ho na yɛne twe nkɔmɔ a |
agric_00000_000039 | 39 | 30 | 3,001 | 154 | -0.6059 | district bamboi baba tour na yɛbɛtwetwe nkɔmɔ ɛha na yɛde mosi nyame yɛ adom me da foforɔ nkɔtwetwe bi a ɛwɔ adwumakuo adwuma tu mpo na ɛkɔ dwumadie yɛde bɛbrɛ mo y'akra mo nananom akuafoɔ hwan biology |
agric_00000_000041 | 41 | 30 | 3,001 | 185 | -0.8613 | sɛn ɛna wo to ɔmofo no atumi ayɛ yie wei deɛ y'anya wo wɔ fie nti nigeriafoɔ monnyɛ no biom kwan bɛn na wofaso etu afu na yɛn nso yɛ fie ha sei nkran rush awakyerɛ sɛ obiara bɔ wo ho mmɔden kwan bɛn na wonfa so aboa yi ɛkwan a mɛfa sole of africa |
agric_00000_000042 | 42 | 30 | 3,001 | 149 | -0.8392 | so ɛberɛ yi ɛyɛ berɛ a ɛɛ nnoɔma ɛnkɔ yie nanso sɛdeɛ meka no daa ɛwɔ adeɛ a ɛhyɛ me yi ho no kyerɛ sɛ nnoɔma nyɛ nanso ɛyɛ dɛn a yɛbɛdidi nti youtube a ɛkɔ adwumadie no ministry of agriculture |
agric_00000_000043 | 43 | 30 | 3,001 | 295 | -0.7776 | cooperative no nti kɔpetifo no ɛmu nkyerɛkyerɛ ɛna ɛkɔ political no ara nso yɛn nso yɛboaboa yɛn ho yɛ kɔ yi ha na yakɔboa no ɛkɔ yi ha na y'akɔboa no nhyehyɛ yɛwura mu yɛ kɔ wo a wodin yaa na yakasa ho akyerɛ wo na yakyerɛ wo kwan a wobɛfa so a wobɛyɛ cocoa no nso yɛmmra mo nsaase no so kakra agyafoɔ yɛbɛdi nkɔmmɔ bio... |
agric_00000_000044 | 44 | 30 | 3,001 | 232 | -0.6752 | sɛ ɛnyɛ nokware kwa na mmom ɛwɔ nnoɔma bi nti na area biara ne adeɛ a wo duru a ɛkɔ yie no infact wo hwɛ aman mu ha na yɛwɔ baabi a ɛyɛ a predominantly no yɛ dua number yɛ bi ɛnyɛ bachance okay usually ɛhɔ asaase no sɛdeɛ asaase no si teɛ ɛɛm asaase no sɛ ɛyɛ ɛyɛ plane ɛda hɔ level anaasɛ mopɛ wɔ mu ɛne |
agric_00000_000045 | 45 | 30 | 3,001 | 92 | -1.1337 | okay great ma memmisa wei ɛɛ yɛreka yɛka adebɔneyɛfoɔ ba mu no yɛmfa no sɛ koko brd ɛyɛ cocoa health development |
agric_00000_000046 | 46 | 30 | 3,001 | 189 | -0.3581 | detwa aduro yɛde fa aduro adi gu afuo soɔ anaa afei w'aburo yi wo susu a sɛn na ɛbɛkɔ n'akwan bɛn koraa nso na wowie a wotɔn no wode sie anaasɛ wobu yɛ a na ɔretutu akɔtɔn aburo eka bɛyɛ ɛnan wode w'ani so a wobɛnya bɛɛma wayɛ sei amanfoɔ bɛwe bi |
agric_00000_000047 | 47 | 30 | 3,001 | 159 | -1.0932 | bɛtena so a ɛtumi hunu sɛ no na ɛno deɛ wakyerɛ mu but memmisa no sɛ sɛ meyi wie na ebia oh mayɛ busy megya bi sei ano ha for baabi a three four hours nti yɛkɔ duru ha kyerɛ sɛ ɔmo ayɛ ready wɔ hɔ a ɔmo bɛ start |
agric_00000_000048 | 48 | 30 | 3,001 | 113 | -0.6536 | adwuma no ayekoo yadwuma no sɛ wayɛ na watumi de bi akɔ ɔman foforɔ so na ɔman no dan wo ɛno ne sɛ the ss no ghana coco ɛna ɛno deɛ yɛ dua a yɛdi bi wɔ |
agric_00000_000049 | 49 | 30 | 3,001 | 174 | -1.028 | da mo ase o akuafoɔ yɛda mo ase abusuafoɔ da hwan wɔ asante akyem agogodwa so ɛna yɛwɔ yi na yɛne agriga director yɛnhwɛ sɛ sɛ yɛde kɔ amanɔne ɛno firi agogo ha no ɛho nsɛm ne adeɛ sɛ figures no koraa te sɛn na yoo medawoase |
agric_00000_000050 | 50 | 30 | 3,001 | 183 | -0.7065 | na wofa boa a nti wo nnɔbaeɛ no fa ebi wo gu so aburo ɛntoose beans ne ne nkeka ho ɛbɛboa wo sanso ɛna wotwe cocoa no aba fie a adwumakuo a yɛfrɛ ɔmo strong sɛ aboboyaa no ara pɛ bi ɛno na ɛbɛtwe de aba fie de ama wo na nnoɔma akɔ yie adwuma |
agric_00000_000051 | 51 | 30 | 3,001 | 226 | -0.5034 | wɔ ha yi wei deɛ ɔfiri ghana ha yi a akyekyedeɛ wei nyɛ ghana deɛ ɛyɛ west africa giant tortis a ɔfiri malisa ba ɔyɛɛ dɛn na dɛn na wo nsa aka ahyehyɛdeɛ sei yɛwɔ ɛɛ mmoa bi wɔ hɔ a wotumi fa na wode afiri kuro foforɔ so na woba a yɛwɔ mmara ɛwɔ ho a wobɛtumi de aba yɛ man mu nti wofa saa kwan no so |
agric_00000_000052 | 52 | 30 | 3,001 | 323 | -0.9049 | ɔmo ate awie baabi abɛyɛ baabiara ɔmo siesie no akɔ akɔtɔn yɛne deɛ ɔda ɔmo ano ɛtwetwe nkɔmɔ na yɛnhwɛ sɛ ghana yɛrekɔ y'anim ɛwɔ kwan a ɛtɔn yɛ nneɛma ho anaa madam mepa wo kyɛw wo din no sɛn mepa wo kyɛw yɛfrɛ me saa akwadaa ten auntie sarah sɛ yɛte aduaba no anaa no baeɛ no bɛduru ha a ɛbaa mo adwuma no sɛ ɛnfa yɛ ... |
agric_00000_000053 | 53 | 30 | 3,001 | 200 | -0.7282 | adwuma mu paa frɛ yɛ a mo baeɛ yi mo ne akuafoɔ no na anka w'atwetwe ne te sɛn yoo anim medaase yɛne akuafoɔ no na nkɔmɔ twetwe no ɛsɔ ma nipa y'akyerɛkyerɛ ɔmo sɛdeɛ yɛ use nnuro ɛne sɛ deɛ yɛ use fatliza ɛne machines ne biribiara mprempren which is organic fa |
agric_00000_000054 | 54 | 30 | 3,001 | 154 | -0.7376 | obiaa biribi nti ama abaayesfoɔ no aba dodoɔ sɛ obiara pɛ bi akɔ india ne rebert sɛ yɛn ghanafoɔ no yɛyɛ ntɔ nti yɛntɔ yi deɛ a ɔmo nso na woba ɔmo de sika ba yi na ghana goverment price service |
agric_00000_000055 | 55 | 30 | 3,001 | 194 | -0.331 | nipa baako ɛkɔfa problem ba a deɛ ɛbɛka ne sɛ hatresi nyinaa ɛnyɛ nanso na ebia ɛnte saa enti deɛ yɛ deɛ yɛ produce kosua no yɛ de no deɛ vaccine ne nnoɔma a wode gu ne mu no wode bɔ ne maame ne ho ban na yɛn nyinaa yɛde ama antibodies bi gu kosua no mu |
agric_00000_000056 | 56 | 30 | 3,001 | 135 | -0.3957 | yɛsrɛ mo okey na ma no bɔ akɔ soro ɔmmeeɛ yɛnni baanu nyinaa yɛngyina ntɛm nyame gyina hɔ ntɛm na yebesi a yɛbɛyɛ me sɛ sesei de kyerɛ sɛ wɔka a na matete wei mpo me ho hwee nanso |
agric_00000_000057 | 57 | 30 | 3,001 | 258 | -0.5311 | kakra a ɛwɔ mu ne sɛ me deɛ mede me nsa ato mu awie na mehwɛ na manya afiri mu nso so a sɛbe ɛnyɛ nipa nyinaa na ɛnyɛ no saa nti mentena ha ɛnyɛ nyame bɔne ayɛ na menyi ghana nso so ɛbone ayɛ na ɛsan sɛ mayɛ adwuma no ahodoɔ bebree enti sɛ baabi ɛrekɔ fam a deɛ ɛbɛyɛ ne sɛ baabi a ɛhɔ ɛrekɔ soro no na mede hɔ no aba ab... |
agric_00000_000058 | 58 | 30 | 3,001 | 197 | -0.6473 | bi na yie na y'ahwɛso sɛ ministry no nso nhyehyɛɛ bi a ɔde baabiara no so no ɛsɛ ɛda nokware no ah na ne kwan mu a sɛbe mu bi deɛ sɛ wonim sɛ ɛnyɛ adeɛ a yɛbɛte aseɛ because yɛda nnipa no ɛna nipa nso na yɛrekasa ma ɔmo ɛna nipa na ɛyɛ kasa ma ɔmo no nso so wɔ expe |
agric_00000_000059 | 59 | 30 | 3,001 | 176 | -0.5754 | cocoa no a afei yɛrekɔ no afei na yɛbɛdi cocoa no mu dɛ nti mesrɛ kwani biara sɛ ɔmo ho mmɔden sɛ wo kɔ kɔ aka nwuram koraa seesei hu seesei mu no sɛnea ɛtete no wɔ mfiri no so no sɛ onyamena adom koraa afei yɛbɛkɔ yɛn anim enti |
agric_00000_000060 | 60 | 30 | 3,001 | 179 | -0.6984 | oh nyame adom ɛyɛ ara because wohwɛ a ɛnam the quality of x a produce wɔ ha nti no yɛnni complete atɔfoɔ mmere pii no mmaa no koraa moba a na ɔmo because wɔn nyankopɔn anaa sɛ like ɔmo star thousands no ɔmo nya no saa nka ɔmo pɛ but |
agric_00000_000062 | 62 | 30 | 3,001 | 179 | -0.5446 | sɛ wobɛtɔn aba anaa sisi aka ho a wo saa hia tu me krataa ɛfiri bprs hɔ enti nneɛma na wɔtɔn no ansa nipa no bɛtumi adi saa adwuma no nkrataa mmiensa na ɔhia a bra no bra no yɛ nyinaa me a mekɔ field but me me nsa question wei nso kwa |
agric_00000_000063 | 63 | 30 | 3,001 | 170 | -0.8395 | mereka 22 24 cocoa sindicated low no ɛbɛduru cocobod kofɛsɛ 27 december a na yɛbue season september enti september october november basically december na sika nni ɔmo de tɔ cocoa enti aka nana stand but ɔnifoɔ 25 y.b |
agric_00000_000065 | 65 | 30 | 3,001 | 200 | -0.7508 | stains no ɛkabom ɛhyɛɛ bi soils a ɛwɔ saa area hɔ ne but yɛnteaseɛ about the liza no ɛwɔ fam every year biara yɛyɛ demonstrations across the country we tell farmers sɛ felises ɛnyɛ adeɛ a yɛba but ɛwɔ sɛ yɛtumi nya the ride fertilizer na yɛtumi ma nnɔ |
agric_00000_000066 | 66 | 30 | 3,001 | 159 | -0.4968 | ahwɛfoɔ a sɛn na woretumi aka sɛ baabi a yɛgyina ha sei sei a no ɛnyɛ amanfoɔ nyinaa ni ei ebi wɔ anim hɔ ɛyɛ me sɛ moafa hɔ but yɛ nyame wie hɔ no deɛ no nti ɛha na yɛahyɛ aseɛ a yɛretim nti na wo no ɛhɔ no nso bɛyɛ |
agric_00000_000067 | 67 | 30 | 3,001 | 270 | -0.4576 | no ɛnyɛ obiaa na ɛde maa wɔn na ɛde saa sika no kɔtoo adb sɛdeɛ ɛbɛyɛ a ɛmu dwuma nkoaa na yɛde yɛ enti no ɛbɛka emufoɔ no nyinaa me ne wo wie w'anom asɛm bɛtwa toa de ma ghana akuafo tituo no ma mo dua mu now ɛsɛ sɛ ɛmo yɛ potential paa na deɛ ɛbɛboa no ne sɛ yɛnni deɛ yɛdua wɔ ha na ɛbɛboa na abaa no nso ɛyɛ ne fa ɔb... |
agric_00000_000068 | 68 | 30 | 3,001 | 194 | -0.6297 | ɔbɛtu na obi nso wo dua na bi rebɔ spray na ebi nsoso ɛdɔ aburoo no na ebi dua enti sɛ saa dwuma wei wotumi yɛ ne nyinaa ayɛ no yie a ghana head paa akua adwuma koraa ayɛ dɛ efisɛ first na yɛ maame no yɛ papa ɔmo de sekan nkoaa na ɛdɔ nti adwuma yɛnnim sɛ |
agric_00000_000069 | 69 | 30 | 3,001 | 190 | -0.7681 | mhm ɛna me deɛ ɛgu mu okay na mede efi nkyerɛkyerɛ okey abusuafoɔ kwan no ara na ɔgu so no na ɛkyerɛ sɛ aburo no ebia watutu deɛ ɛnnɛ watutu bi a ɛnora anaasɛ ne nyinaa ɔmo atutu no aburoo dua ɛfuo no ɛso bɔtɔ a ɛwom peinspectors no nyinaa ɛwɔ sɛ ɛ |
agric_00000_000070 | 70 | 30 | 3,001 | 207 | -0.6124 | spirit no ɛgu mu ɛneɛ aduro biara a wode spray biara no but asɛ ɛyɛ ghana ha no nso yɛwɔ aduro a aban akyerɛkyerɛ sɛ yɛmfa ɛnyɛ ɛyi kyerɛsɛ slaa aduro nyinaa akɔ tafo ama wɔn attest ama ɔmo ahunu sɛ ɛyɛ good nti sɛ wobɛtɔ aduro a ɔde bɛham nti wei firi ha nom no na ɛre |
agric_00000_000071 | 71 | 30 | 3,001 | 257 | -0.3382 | antkaa yɛn kɔkɔyɛ ne adeɛ no ɛbɛko n'anim ntɛm so na anka adeɛ yɛka wɔ ghana no ɔno yɛ bi na ɔne amfɛmfoɔ yɛ bi na ebi yɛn na ɔkɔgye bɛkum na ɛma yɛnam awie no sika saa n'adwumadie no na ɛhyehyɛ no toa so a anka ɛbɛkɔ n'anim saa nso ɛna nkosua bɛka bere bi yɛne no nkɔmɔ twetwee mu no kwan a yɛbɛfa so a nkosua koraa no ... |
agric_00000_000072 | 72 | 30 | 3,001 | 117 | -1.4035 | agye ato mu afiri ghana nti ɔmo a ɔmo redi dwa kɔ abɔnten no ɔmo mmra na yɛnyɛ nhyehyɛ ghana yɛgye wodi a ɛbɛyɛ yie o mo mmiensa mu na ɛnyɛ mahunu potential |
agric_00000_000073 | 73 | 30 | 3,001 | 194 | -0.7702 | fine nea ɛnyɛ ara ne sɛ baabia nsuo tae deɛ ɛno deɛ wode yɛ pɛ ɛbɛporɔ okay yɛnhwɛ na afei fa yɛtoa so sɛ yɛdi buronya wie a na yɛse january mu na kyerɛ sɛ afe no akɔ awie fi saa berɛ no ara ɛkɔ afe no awie berɛ bɛn na ɛwɔ brɛ aseɛ okay na saa time no ɛbɛdu |
agric_00000_000074 | 74 | 30 | 3,001 | 136 | -0.6157 | ha yi no na ɛbɛn ne ti no nti wode wo wɔ ha sei na aduro no firi ha nom a na ɛyɛ adwuma aba ne ti no mu sɛ nnɛ wobɛnya atenka no ahaa wode wo ha yi na ɛbɛkyɛ wode wo hɔ na ɛdi 829 hours ansa |
agric_00000_000075 | 75 | 30 | 3,001 | 145 | -0.5896 | ɛwɔ kwan a akuani no bɛfa so wɔ ne extension office a ɛwɔ hɔ no bɛdi nkitaho akuafoɔ no bebree wɔ hɔ a extension officers no deɛ problem ɛwɔ sɛ ɔmo como problem ɛna coco polls no wokɔ |
agric_00000_000076 | 76 | 30 | 3,001 | 126 | -0.5637 | yɛbɛn na ɛwɔ ha nti a ghana asaase so yɛwɔ srɛ mu abaa bɛka nkran dabɛn na ɛwɔ agogo a ɛyɛ soronko a ɛma nantwi yɛn no ɛbɛgye ntena ha wɔ ha ɛdeteɛ ne sɛ deɛ follow |
agric_00000_000077 | 77 | 30 | 3,001 | 185 | -1.0924 | okay ɛnoara bɔ life ɛno ara bɛbɔ line na two tokura ama wo na wode buro no agugu mu okay na wei nso dwuma bɛn na ɛdi wei nso yɛ tractor ɛno nso yɛ the same principle okay ɛno wei no wodi hyɛ tractor no rekyia anaa track no ɛtwi kyerɛ sɛ wei di dw |
agric_00000_000078 | 78 | 30 | 3,001 | 217 | -0.7432 | tiri yɛn wo mmara yɛanhwɛ no yie a daakye yɛn ho bɛkyere yɛn nkyiri mma no wobɛkyerɛ wo papapapa mmara no nso yɛyɛ a ɛɛ wonhu no yie woahu no yie me deɛ mehu no yie because yɛse asɛm mmra sɛ wobɛkɔ na community no gold me yɛtu ne galamsey wokɔka a ɔse ɔmo de america ne community |
agric_00000_000079 | 79 | 30 | 3,001 | 181 | -0.7204 | wei deɛ na yɛfaa yɛkɔ adwumadie so na yɛahyɛ of agric ɛna environmental protection agency yi epa sɛ mommra mu esiane sɛ sɛ wosɔre na woanya biribiara nni a abrabɔ bɛyɛ den ama wo na sɛ obi bɛtu anwea na ɛkɔm bɛde ghana ma deɛ ɛmuɔ |
agric_00000_000080 | 80 | 30 | 3,001 | 250 | -0.7385 | ɛwo mu ara na bi memfa no sɛ abofra na na yɛdua kakra kakra ɛka ho ogya bɛtɔ asaase no adɛn na ɛwɔ so nsan sɛ deɛ nti na me bisa wo saa no sɛ ghana yɛn nsa ase na sɛ agya ewiem bɔbrɛ a ɛsesa nti no bi bɛyɛ kuadwuma a tesɛ deɛ irrigation a akyire yɛbɛdi ho nkɔmmɔ yi na ɛno akɔ akɔ ba mu asaase na n'abakwasɛm te sɛn kakr... |
agric_00000_000081 | 81 | 30 | 3,001 | 163 | -0.9121 | saa okay wo dua green a wonnya no saa okay sɛ obiara nnua seed no nea woaka yi under planting for food mennyɛ sure sɛ ɛyɛ bronkoaa nti wei deɛ mayɛ kakra nneɛma bɛn na ɛwɔ sɛ yɛkɔ gye seed no aburo pɛ baako ɛyɛ |
agric_00000_000082 | 82 | 30 | 3,001 | 169 | -0.8494 | sɛn medawoase ɛyɛ me sɛ wan for life finishetive a ɛyɛ yɛ founder doctor kwabena agyei anisoadehunu a md so mr rekyerɛ gyei ɛɛ wodi anim a yɛ make sure sɛ at least i will prefer drink water kaeɛ na ɛnyɛ asɛm ketewa ɛɛ |
agric_00000_000083 | 83 | 30 | 3,001 | 247 | -0.4581 | sene ɛwɔ ɛkwan ahoroɔ bebree so na mmom yɛn five no yeah seesei nea ɛyɛ ne sɛ biribiara ni a wofrɛ a ɛsi regional office anaa zonal officer no so a na wama yaso ase na yɛn nso nea ɔbɛn wo no na yama no na hɛɛ ɛnhwɛ anidie but yɛrebɔ yɛ ho mmɔden na yɛnim sɛ mmerɛ pa mu no aban nso ɛbɛboa ama nipa no bebree na abɛka yɛ |
agric_00000_000084 | 84 | 30 | 3,001 | 159 | -0.3676 | woyi gusakuzafoɔ na ɛbaeɛ no wo yɛ adwuma yi accounting adwuma na woyɛɛ no ka sɛ adwuma deɛ yɛyɛ aso mfasoɔ anaasɛ yɛyɛ anya sɛdeɛ asi ayɛ no nkontabuo waduane tɔ nsuo ne akorakora ne adeɛ no kwan bɛn na |
agric_00000_000085 | 85 | 30 | 3,001 | 186 | -0.591 | paa obiara te ne nka nti mede meho kɔhyɛɛ kwaadwuma mu na na m adɔ kakra bɛyɛ sɛ 10 years wɔ hɔ dada nti deɛ meyɛɛ ne sɛ ɛneɛ mede me sika ma yɛn nyinaa bɛ gu so enti first year no deɛ ɔbɛyɛ 40 acres bom ɛyɛ cocoa cocoa bɛyɛ antɔ deɛ no because na me |
agric_00000_000086 | 86 | 30 | 3,001 | 174 | -0.5855 | a wonnya nti nea ɛyɛ a yɛreyɛ ne sɛ aburo no a yɛ dua koraa no wobu aburo no wie a n'aba a na ɛsi hɔ no ɛyɛ ne stok no ne ne nwura no nyinaa ɔmo wei nti wowɔ baabi a ɔmo nso adua aburo ne ya ɔmo twa ne fresh no a wobɛtumi agye na wode akɔ |
agric_00000_000087 | 87 | 30 | 3,001 | 259 | -0.6723 | doctor asamoa ho dɔ mensaa ce of salmon farms yɛbaa ha na kube yi ɛyɛ nketenkete wahu sɛ apɛgya paa adwuma a ɛkɔ sɛn oh nyame adom a ɛkɔ yie paa biribi foforɔ bi wɔ hɔ a dwuma yi wode aka ho sɛ phone adwuma sɛ ebia kube y'atwa me amanfoɔ but abusuafoɔ deɛ ɛwɔ mu ne sɛ mpuntuo aba no sɛ kube no yɛbaa first one yi kyerɛ ... |
agric_00000_000088 | 88 | 30 | 3,001 | 114 | -1.5345 | ghana akuafo ɛyɛ akuafoɔ yɛda mo ase nyansa baako ano a akuafoɔ eee ohh ayekoo ghana akuafo emu ayɛ bi oo nananom akuafo emu afia ei deɛ mpo nya |
agric_00000_000089 | 89 | 30 | 3,001 | 296 | -0.7543 | yɛwɔ west africa center for crop employement legon aha ɛna yɛ ne nimdefoɔ no bɛtwetwe bɔne ahwɛ abɛɛfo kwanso sɛ yɛpɛ seed deɛ yɛbɛte sɛ cd foforɔ aba henfa ɛno na ɛnnɛ yɛne nimdefoɔ ɛreba abɛtwetwe ho na abotare na tena ase na tie bi na wonya biribiara wonhu sɛ anamon tuo no ɔman no gu so reto baabi a wose wohunu no n... |
agric_00000_000090 | 90 | 30 | 3,001 | 250 | -0.5849 | a ɔbɛ nenam hɔ saa time no ayɛ sɛ woto ma wo no ɔnto mma wo but two aduane a ɛyɛ sɛ ɔmo de ma ɔmo akokɔ biara ne kwan tete a ɔfe da wode ma no nti w'anya da right quantity wɔ fie da obedi ɛno nti ɔntumi nyin na ɔnto mma wo medication nnuro a ɛsɛ wode ma ɔmo anfa ɔmo vaccination saa nnoɔma nyinaa woanyɛ no pɛpɛɛpɛ amma ... |
agric_00000_000091 | 91 | 30 | 3,001 | 116 | -0.755 | yɛtwetwe nkɔmmɔ na ɛnnɛ deɛ agya ɛyɛ me sɛ tɛkrɛma no adane kakra yɛwɔ mframa ntɛm mu owura benkam mm ɔbaa ghana bɛyɛ kwaadwuma edu baabi a yɛnansa ɛ |
agric_00000_000092 | 92 | 30 | 3,001 | 107 | -0.6345 | tumi te sɛ ɔmo tumi nya nam na ɔmo ayi bebree koraa na afei abu so wɔ mu asan agya bi agu nsuo no mu ɛberɛ bɛn na saa adeɛ no ɛba ɛɛ first okay |
agric_00000_000093 | 93 | 30 | 3,001 | 169 | -0.5412 | akoko kakra a medidi so na ɛbɛba sɛ yɛbɛ twa agu ɛfiri sɛ cocoa krakyefoɔ no ba mfuo no nyinaa ɔmo maame teaseɛ sɛ baabi a megye di sɛ ɛhɔ koraa na menyaa ba kakra afiri no koraa no yareɛ no ade koro hɔ no nyinaa nti yɛtwam |
agric_00000_000094 | 94 | 30 | 3,001 | 138 | -0.834 | a ɛtaa ha mo adwene ɛɛm abɛ abɛ a ɛwɔ hɔ yi deɛ yare bebree nta nha no the main issue ɛha abɛ wɔ ghana yɛ pests ɛɛ mmoa bi a yɛfrɛ mo leave mining wɔ northern ɛhwɛ every two weeks bi |
agric_00000_000095 | 95 | 30 | 3,001 | 143 | -0.4274 | adv afa ne ntampa afura wo kwan bebree do ɛkaa nom ɛna yɛyɛ no ɛna y'ase sɛm ɛnnɛ yɛ pia kuafoɔ no sɛ ɛbɛhwɛ a yɛdze branches aba yɛn mmra yɛdze aba yɛn branches bebree ɛna afei nsoso |
agric_00000_000097 | 97 | 30 | 3,001 | 135 | -1.4365 | anaa in the wake of climate change challenges okay online those days wedding will live the remains the rest of the field no we leave them on the field alright before |
agric_00000_000098 | 98 | 30 | 3,001 | 266 | -0.4915 | okay nkɔmmɔ na ɛdɛ na yɛbrɛ no nti yɛnhwɛ obi pɛ sɛ ɔsua ayekoo deɛ y'adwumadie ne yɛdaworoma sɛ wode ayɛ yie no na obi rehwɛ sua ɛyɛ dɛn a makyerɛ na ɛbɔɔ ne hɔ sekan me brɛ sɛ na waka me yie ho ka ne ɛberɛ no so te sɛn cos obi wɔhɔ sɛ woyɛɛ da koro pɛ na ɛyɛ yie ɛho ka ɔbrɛ ne ade abrɛ wo ho so na ɛte sɛn koko dwuma ... |
agric_00000_000099 | 99 | 30 | 3,001 | 151 | -0.3297 | a 150,000 no so no party biara wɔ deɛ ɛrekɔ yɛwɔ baabi koraa ɔno nam no mu a nkwadaa sɛbe ɔmo adwene wɔ fam no yɛde yɛ bɛbi food mu a ɔmo di ama wɔn adwene ebue na mpo koraa wanya kata ano yɛ den so mfaso koto |
agric_00000_000100 | 100 | 30 | 3,001 | 136 | -0.8559 | yi a me dwumadi a mereyɛ no no afe 2020 but a wofive mo ne ghana akuafoɔ deɛ wodi agorɔ deɛ dua mu dodua cocoa deɛ duru ayuo ɛkɔɔ sɛn medaase afe 295 nokwasɛm yɛ hyɛ akuafoɔ agye y'asoma |
agric_00000_000101 | 101 | 30 | 3,001 | 233 | -0.7501 | deɛ nti ɛyɛ a na yɛnya dɛn inflammatory ɔmo ahunu sɛ ebia 1.8 no koraa afe wei fulani mu sesia adeɛ yenni tracking system but afei na yɛreyɛ saa nhyehyɛeɛ yi 100 transit months problem ahaa nti nhyehyɛeɛ ne wo ba sɛ aboa bi refi reba a yɛtumi bɛ track ɔwɔ baabi a ɔbɛfa ɔbɛfa kwarantan yɛbɛ twerɛ ne din de |
agric_00000_000102 | 102 | 30 | 3,001 | 169 | -1.0096 | wo ho yɛ yi saa time no na ne ba na ɛnyie ama m'atɔn ɔmo ama wawo biom atɔ wo wɔ afei na wobɛyɛ mprensa enti sɛ frigo no ara wɔ 1414 mmiensa a na metɔn baako biara tenaa ne tenaa na ɛda tams 14 time three a na kyerɛ sɛ metwa nantwie |
agric_00000_000103 | 103 | 30 | 3,001 | 147 | -0.4921 | kɔ ɛɛ ɔbaa pond baako w'anya baako baako nsia ɛnnɛ koraa deɛ physical deɛ ɛbɛ sua adeɛ wɔ ha yɛnwura mu kakra wode nam no gu nsuo mu ne adeɛ no ayekoo deɛ wahwɛ no hyew baako kɔ ɛho korako nti ɛ |
agric_00000_000104 | 104 | 30 | 3,001 | 198 | -0.9416 | na wode asepɛ no anaa sɛ nkran na ɛtɔ car soɔ no but complains aba sɛ ebinom ɛba ha nom ɛbɛtɔ na menim sɛ ahyɛ aseɛ mfie sɛ biribi saa wɔ hɔ koraa hyɛ asɛm bi a ɛrekɔ so saa wahu sɛ mmara biara wo start na ɔyɛ employment yɛrebɔ mmɔden biara sɛ mmara tcba |
agric_00000_000105 | 105 | 30 | 3,001 | 166 | -0.7336 | nea eye no yɛnam ne kwan so pɛpɛɛpɛ na yɛreyɛ yɛn ankasa yɛnfɔɔso nka yɛ tɔn euro baabiara nti me nua ɛnim sɛ wobisa asase mu a nnoɔma a y'ayɛ a ɛde abɛduru saa pɛpɛ soɔ no ɛna maka akyerɛ wo abusuafoɔ asɛmpatia |
agric_00000_000106 | 106 | 30 | 3,001 | 191 | -0.4043 | nea owie a ɔbɛhwɛ na waboaboa no nea ɔbɛtɔ ɛne nea ɛsɛ sɛ ɔdanedane no enti yɛde ne nyinaa ka bom maame medaase mɛkɔ me nua hɔ na m'ase aba wo hɔ biom na ɛyɛ bra baako wɔ ha edi kan ɛmmisa ne din no afei ɔno nso ghana rice project no a agra ɛreboa a |
agric_00000_000107 | 107 | 30 | 3,001 | 119 | -0.6543 | nti kumasi ɛna mekɔ kɔtena kakra na ɔse ɔma meyɛ car adwuma nti bra me mede car no kɔɔ hɔ no na wofrɛ no agyae gyae nti agya ba no papa ɛnyɛ baako nti woyɛ agya |
End of preview.
Twi Grapheme-Unit Feature Store
The training data behind ghana-pico-asr: 805 hours of Twi speech turned into log-mel features with a grapheme-unit label for every 10 ms frame, produced by CTC forced alignment. Publishing it means the expensive step -- aligning 805 hours on a GPU -- does not have to be repeated to train, reproduce or extend the model, and the same pipeline can be pointed at a new language.
This is a derived feature store, not a speech corpus: it contains 40-band log-mel spectrograms, not audio. The original recordings live in the source datasets listed below.
What a label is
Twi is written with digraphs that behave as single sounds (ky gy hy ny tw dw kw gw hw nw), so the label alphabet is grapheme units rather than letters:
each unit is one of those digraphs, a single letter, or <sil>, for
37 classes in total. Text is segmented greedily, longest match first,
so akyekyedeɛ becomes a ky e ky e d e ɛ.
Alignment used MahmoudAshraf/mms-300m-1130-forced-aligner at a 20 ms emission stride,
with a <star> wildcard allowed at utterance edges and word boundaries only --
never inside a word, where it would happily swallow a real unit.
Contents
| path | what |
|---|---|
features/mel/<source>_<shard>.npy |
float16 log-mel, 40 bands, 10 ms hop, 16000 Hz, one blob per shard |
features/manifest/<source>_<shard>.npz |
per-utterance unit ids, frame spans and alignment scores |
features/manifest/<source>_<shard>.jsonl |
the same, human-readable, one row per utterance |
features/vocab.json |
unit -> class id; the ordering is part of the model contract |
The per-frame label tracks and chunk index are not shipped: the training code builds them from the manifests on first run and caches them, and because chunk assignment comes from hashing utterance ids, the train/validation/test split it produces is identical to the one the released model was trained on.
Scale
| audio | 805.04 h |
| utterances | 294,005 |
| shards | 313 |
| 1 s chunks (500 ms stride) | 5,671,770 |
| labelled frames | 173,562,329 (59.9% of frames) |
| classes | 37 |
The remaining frames are deliberately unlabelled (IGNORE_INDEX) and
contribute no loss: inter-unit gaps too long to attribute to either neighbour,
and gaps that are loud enough that calling them silence would be a lie.
<sil> is 33.0% of labelled frames.
Sources
Numeric ids, so the data does not hard-code corpus identities and stays valid if a corpus is renamed. Ids are append-only.
| id | dataset | transcripts |
|---|---|---|
| 1 | ghanaopenai/new-twi-tts-aligned | human |
| 2 | ghananlpcommunity/twi-health-asr-gemini-500hrs | machine (Gemini) |
| 3 | ghananlpcommunity/kumawood-speech-transcriptions | machine (Google STT) |
| 4 | ghananlpcommunity/ghana-female-twi-asr-16word-splits | human |
| 5 | ghanaopenai/twi-agriculture-speech | human |
| 7 | ghanaopenai/asante-twi-bible-speech-text | human |
Id 6 (ghanaopenai/twi-speech-text-multispeaker-16k) is reserved but absent:
its audio could not be decoded. A minority of shards from ids 1 and 4 are
likewise missing -- the store holds what aligned cleanly, and shard names are
therefore not contiguous.
Utterances were kept only at 0.4-31 s, at least 3 units, and 3-32 units/second; a rate outside that band means the transcript and the audio disagree, which is a real risk with machine transcripts.
Using it
from huggingface_hub import snapshot_download
snapshot_download(
repo_id="ghanaopenai/twi-grapheme-unit-features",
repo_type="dataset",
allow_patterns=["features/**"],
local_dir="/data", # then train against /data
)
Training, evaluation and inference code, including the alignment stage that produced this store and the language registry for adding another language, is at https://github.com/ghanaopenai/ghana-pico-asr.
Provenance and licence
Each source corpus keeps its own terms; consult the datasets linked above before redistribution or commercial use. No audio is included here.
- Downloads last month
- -