Datasets:

sulabhkatiyar
/

ne-asr-dataset-njo-aug

text stringlengths 27 198	language stringclasses 1 value	augmentation stringclasses 3 values
arudang sensaktsüba atema lir aser idakji bottle a ka lir	Ao	original
arudang sensaktsüba atema lir aser idakji bottle a ka lir	Ao	speed_0.9
arudang sensaktsüba atema lir aser idakji bottle a ka lir	Ao	speed_1.1
iba noksa nung ya tzuyim tulu ka nung patak ana senzua lir	Ao	original
iba noksa nung ya tzuyim tulu ka nung patak ana senzua lir	Ao	speed_0.9
iba noksa nung ya tzuyim tulu ka nung patak ana senzua lir	Ao	speed_1.1
cari yanglutsu azük nung yua lir ano iba tepok nungji nisung ka mena angur	Ao	original
cari yanglutsu azük nung yua lir ano iba tepok nungji nisung ka mena angur	Ao	speed_0.9
cari yanglutsu azük nung yua lir ano iba tepok nungji nisung ka mena angur	Ao	speed_1.1
table ka nung nisung ka mena lir table ka nung nisung asem mena lir	Ao	original
table ka nung nisung ka mena lir table ka nung nisung asem mena lir	Ao	speed_0.9
table ka nung nisung ka mena lir table ka nung nisung asem mena lir	Ao	speed_1.1
kulen nisung ka chair nung mener repranga lir kulen nisung ka nokdaker repranga lir	Ao	original
kulen nisung ka chair nung mener repranga lir kulen nisung ka nokdaker repranga lir	Ao	speed_0.9
kulen nisung ka chair nung mener repranga lir kulen nisung ka nokdaker repranga lir	Ao	speed_1.1
anasa süngdong ana lir aser tzülu nung tzüa lir aser iba jak tetong ya green colour lir	Ao	original
anasa süngdong ana lir aser tzülu nung tzüa lir aser iba jak tetong ya green colour lir	Ao	speed_0.9
anasa süngdong ana lir aser tzülu nung tzüa lir aser iba jak tetong ya green colour lir	Ao	speed_1.1
aser süngdong nungera tapu balala angur	Ao	original
aser süngdong nungera tapu balala angur	Ao	speed_0.9
aser süngdong nungera tapu balala angur	Ao	speed_1.1
iba photo nung ya table ka angur iba table ya colour tanak nung lir	Ao	original
iba photo nung ya table ka angur iba table ya colour tanak nung lir	Ao	speed_0.9
iba photo nung ya table ka angur iba table ya colour tanak nung lir	Ao	speed_1.1
table ka toko nungbo basket ka lia angur iba basket ya yellow colour nung lir	Ao	original
table ka toko nungbo basket ka lia angur iba basket ya yellow colour nung lir	Ao	speed_0.9
table ka toko nungbo basket ka lia angur iba basket ya yellow colour nung lir	Ao	speed_1.1
iba dokan ya tanur osettsüset ayokba taser parnok kazütsü süoshi nunger ayokba ka lir	Ao	original
iba dokan ya tanur osettsüset ayokba taser parnok kazütsü süoshi nunger ayokba ka lir	Ao	speed_0.9
iba dokan ya tanur osettsüset ayokba taser parnok kazütsü süoshi nunger ayokba ka lir	Ao	speed_1.1
ano ali nung hospital nung amshiba oset kar yua lir	Ao	original
ano ali nung hospital nung amshiba oset kar yua lir	Ao	speed_0.9
ano ali nung hospital nung amshiba oset kar yua lir	Ao	speed_1.1
iba nung ya steel agi yangluba cup nunger jar nunger aika dang yua lir	Ao	original
iba nung ya steel agi yangluba cup nunger jar nunger aika dang yua lir	Ao	speed_0.9
iba nung ya steel agi yangluba cup nunger jar nunger aika dang yua lir	Ao	speed_1.1
iba noksa nung ya chair ka tema süoshi aika dang enoka lir	Ao	original
iba noksa nung ya chair ka tema süoshi aika dang enoka lir	Ao	speed_0.9
iba noksa nung ya chair ka tema süoshi aika dang enoka lir	Ao	speed_1.1
nisung as pezü nokdaka lir nisung anabo mena lir	Ao	original
nisung as pezü nokdaka lir nisung anabo mena lir	Ao	speed_0.9
nisung as pezü nokdaka lir nisung anabo mena lir	Ao	speed_1.1
iba anasa ya stage tilaka nung nisung ka nokdaka lir	Ao	original
iba anasa ya stage tilaka nung nisung ka nokdaka lir	Ao	speed_0.9
iba anasa ya stage tilaka nung nisung ka nokdaka lir	Ao	speed_1.1
pa sülen ji shelf aika dang enoka lir iba shelf nung ya medical oset aika dang enoka lir	Ao	original
pa sülen ji shelf aika dang enoka lir iba shelf nung ya medical oset aika dang enoka lir	Ao	speed_0.9
pa sülen ji shelf aika dang enoka lir iba shelf nung ya medical oset aika dang enoka lir	Ao	speed_1.1
iba noksa ya tsümar puja asüba area ka lir	Ao	original
iba noksa ya tsümar puja asüba area ka lir	Ao	speed_0.9
iba noksa ya tsümar puja asüba area ka lir	Ao	speed_1.1
carpet ji karbo blue karbo red karbo orange ta nung lir	Ao	original
carpet ji karbo blue karbo red karbo orange ta nung lir	Ao	speed_0.9
carpet ji karbo blue karbo red karbo orange ta nung lir	Ao	speed_1.1
iba gate colourji tanak dang agur aserbo ki tilaka indang anasaji yellow colour blue colour agi indang agi süng agi yanglua ngudar	Ao	original
iba gate colourji tanak dang agur aserbo ki tilaka indang anasaji yellow colour blue colour agi indang agi süng agi yanglua ngudar	Ao	speed_0.9
iba gate colourji tanak dang agur aserbo ki tilaka indang anasaji yellow colour blue colour agi indang agi süng agi yanglua ngudar	Ao	speed_1.1
parnok ana anasaji tesayur ana nokdaka lir	Ao	original
parnok ana anasaji tesayur ana nokdaka lir	Ao	speed_0.9
parnok ana anasaji tesayur ana nokdaka lir	Ao	speed_1.1
entrance aoba lenmangji maroon colour tepa lir	Ao	original
entrance aoba lenmangji maroon colour tepa lir	Ao	speed_0.9
entrance aoba lenmangji maroon colour tepa lir	Ao	speed_1.1
tetsür ana mena lir aser tebur ka nokdaka lir	Ao	original
tetsür ana mena lir aser tebur ka nokdaka lir	Ao	speed_0.9
tetsür ana mena lir aser tebur ka nokdaka lir	Ao	speed_1.1
shelf ji temesüng colour frame temji brown colour aser gold colour nung lir	Ao	original
shelf ji temesüng colour frame temji brown colour aser gold colour nung lir	Ao	speed_0.9
shelf ji temesüng colour frame temji brown colour aser gold colour nung lir	Ao	speed_1.1
mountain nunger angur aser building a kar angur	Ao	original
mountain nunger angur aser building a kar angur	Ao	speed_0.9
mountain nunger angur aser building a kar angur	Ao	speed_1.1
ground tuluka lir aser iba ground nungya park tuluka dang yanglua lir	Ao	original
ground tuluka lir aser iba ground nungya park tuluka dang yanglua lir	Ao	speed_0.9
ground tuluka lir aser iba ground nungya park tuluka dang yanglua lir	Ao	speed_1.1
aser iba lake ya surround süa wall yagi fencing oa lir	Ao	original
aser iba lake ya surround süa wall yagi fencing oa lir	Ao	speed_0.9
aser iba lake ya surround süa wall yagi fencing oa lir	Ao	speed_1.1
iba en entrance nungji welcome to innuagural function of echahaba tea factory ta zülua lir	Ao	original
iba en entrance nungji welcome to innuagural function of echahaba tea factory ta zülua lir	Ao	speed_0.9
iba en entrance nungji welcome to innuagural function of echahaba tea factory ta zülua lir	Ao	speed_1.1
notebook balala lir tape nunger ayoker colour nunger ayoker aser calender nungera lir	Ao	original
notebook balala lir tape nunger ayoker colour nunger ayoker aser calender nungera lir	Ao	speed_0.9
notebook balala lir tape nunger ayoker colour nunger ayoker aser calender nungera lir	Ao	speed_1.1
flag tilala enoka lir lenmang nungji süngdong nungera angur	Ao	original
flag tilala enoka lir lenmang nungji süngdong nungera angur	Ao	speed_0.9
flag tilala enoka lir lenmang nungji süngdong nungera angur	Ao	speed_1.1
mountain nunger angur aser kia ka angur iba ki indang roof ji temerem colour aser green colour wall ji temesüng colour	Ao	original
mountain nunger angur aser kia ka angur iba ki indang roof ji temerem colour aser green colour wall ji temesüng colour	Ao	speed_0.9
mountain nunger angur aser kia ka angur iba ki indang roof ji temerem colour aser green colour wall ji temesüng colour	Ao	speed_1.1
tetsür kabo temang tulu laji saree temerem sema lir aser tetsür kaji kurta purple colour sema lir aser phone pink colour ama lir	Ao	original
tetsür kabo temang tulu laji saree temerem sema lir aser tetsür kaji kurta purple colour sema lir aser phone pink colour ama lir	Ao	speed_0.9
tetsür kabo temang tulu laji saree temerem sema lir aser tetsür kaji kurta purple colour sema lir aser phone pink colour ama lir	Ao	speed_1.1
tetsür kati tanur ka ama lir tetsürla ji purple colour scarf agi tekolak shibanga lir aser tanur indang tolaji neon colour lir	Ao	original
tetsür kati tanur ka ama lir tetsürla ji purple colour scarf agi tekolak shibanga lir aser tanur indang tolaji neon colour lir	Ao	speed_0.9
tetsür kati tanur ka ama lir tetsürla ji purple colour scarf agi tekolak shibanga lir aser tanur indang tolaji neon colour lir	Ao	speed_1.1
kima nung süngdong tapu balala lir kiji green colour lir	Ao	original
kima nung süngdong tapu balala lir kiji green colour lir	Ao	speed_0.9
kima nung süngdong tapu balala lir kiji green colour lir	Ao	speed_1.1
pine trees angur naro tapu balala angur	Ao	original

End of preview. Expand in Data Studio

NE ASR Augmented Dataset -- Ao (njo)

Augmented automatic speech recognition dataset for Ao (njo), a Tibeto-Burman language spoken in Nagaland, India.

Source

Augmented from sulabhkatiyar/ne-asr-njo (original transcribed speech data from the ARTPARK-IISc Vaani project).

Language Information

Property	Value
Language	Ao
ISO 639-3	`njo`
Family	Tibeto-Burman
Region	Nagaland, India
Tonal	Yes
Tier	B (0.49h original data)

Dataset Statistics

Original training samples: 259
Augmented training samples: 777 (3x augmentation)
Train shards: 2
Estimated original duration: ~0.5 hours
Estimated augmented duration: ~1.5 hours

Split	Samples
train	777
validation	20
test	10

Transformations Applied

Each original training sample produces 3 samples (1 original + 2 speed + 0 pitch):

Speed perturbation: 0.9x, 1.1x (2 variants per sample)
Pitch shift: Disabled (tonal language -- pitch shift would alter lexical meaning)
Noise augmentation: Not applied

SpecAugment Parameters (for training, NOT in this dataset)

These parameters are consumed by the training script and are not baked into the audio files:

mask_time_prob: 0.05
mask_time_length: 10
mask_feature_prob: 0.03
mask_feature_length: 10
layerdrop: 0.05

Full augmentation config: configs/augmentation_config.yaml

Dataset Format

Audio: 16kHz mono WAV (stored as Parquet with audio bytes)
Text: Transcriptions
Features: audio, text, language, augmentation
Augmentation labels: original, speed_0.9, speed_1.1

How to Use

from datasets import load_dataset

# Load the full dataset
ds = load_dataset("sulabhkatiyar/ne-asr-njo-aug")

# Load only the training split
train = load_dataset("sulabhkatiyar/ne-asr-njo-aug", split="train")

# Filter to only original (non-augmented) samples
original_only = train.filter(lambda x: x["augmentation"] == "original")

# Filter to a specific augmentation type
speed_09 = train.filter(lambda x: x["augmentation"] == "speed_0.9")

Original Data

Source dataset: sulabhkatiyar/ne-asr-njo
Project: ARTPARK-IISc Vaani
License: CC-BY-4.0

Citation

If you use this dataset, please cite the Vaani project and acknowledge the augmentation pipeline.

Downloads last month: 52

Models trained or fine-tuned on sulabhkatiyar/ne-asr-dataset-njo-aug

sulabhkatiyar/ne-asr-njo

Automatic Speech Recognition • Updated 5 days ago • 34