Datasets:
metadata
dataset_name: am-pranav/stt-unified-bench
pretty_name: STT Unified Mini-Bench
task_categories:
- automatic-speech-recognition
task_ids:
- speech-recognition
license: other
size_categories:
- n<10K
configs:
- config_name: assamese
data_files:
- split: val
path: assamese/val-*
- config_name: bodo
data_files:
- split: val
path: bodo/val-*
- config_name: dogri
data_files:
- split: val
path: dogri/val-*
- config_name: en_indian_accent
data_files:
- split: val
path: en_indian_accent/val-*
- config_name: gujarati
data_files:
- split: val
path: gujarati/val-*
- config_name: hindi
data_files:
- split: val
path: hindi/val-*
- config_name: kannada
data_files:
- split: val
path: kannada/val-*
- config_name: kashmiri
data_files:
- split: val
path: kashmiri/val-*
- config_name: konkani
data_files:
- split: val
path: konkani/val-*
- config_name: maithili
data_files:
- split: val
path: maithili/val-*
- config_name: malayalam
data_files:
- split: val
path: malayalam/val-*
- config_name: manipuri
data_files:
- split: val
path: manipuri/val-*
- config_name: marathi
data_files:
- split: val
path: marathi/val-*
- config_name: nepali
data_files:
- split: val
path: nepali/val-*
- config_name: odia
data_files:
- split: val
path: odia/val-*
- config_name: punjabi
data_files:
- split: val
path: punjabi/val-*
- config_name: sanskrit
data_files:
- split: val
path: sanskrit/val-*
- config_name: santali
data_files:
- split: val
path: santali/val-*
- config_name: sindhi
data_files:
- split: val
path: sindhi/val-*
dataset_info:
- config_name: assamese
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
decode: false
- name: text
dtype: string
- name: source
dtype: string
- name: id
dtype: string
splits:
- name: val
num_bytes: 186681046
num_examples: 1000
download_size: 182177487
dataset_size: 186681046
- config_name: bodo
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
decode: false
- name: text
dtype: string
- name: source
dtype: string
- name: id
dtype: string
splits:
- name: val
num_bytes: 169987794
num_examples: 1000
download_size: 166335246
dataset_size: 169987794
- config_name: dogri
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
decode: false
- name: text
dtype: string
- name: source
dtype: string
- name: id
dtype: string
splits:
- name: val
num_bytes: 250328972
num_examples: 1000
download_size: 243221303
dataset_size: 250328972
- config_name: en_indian_accent
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
decode: false
- name: text
dtype: string
- name: source
dtype: string
- name: id
dtype: string
splits:
- name: val
num_bytes: 178921558
num_examples: 1000
download_size: 163516587
dataset_size: 178921558
- config_name: gujarati
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
decode: false
- name: text
dtype: string
- name: source
dtype: string
- name: id
dtype: string
splits:
- name: val
num_bytes: 236196428
num_examples: 1000
download_size: 231117224
dataset_size: 236196428
- config_name: hindi
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
decode: false
- name: text
dtype: string
- name: source
dtype: string
- name: id
dtype: string
splits:
- name: val
num_bytes: 169694428
num_examples: 1000
download_size: 163153426
dataset_size: 169694428
- config_name: kannada
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
decode: false
- name: text
dtype: string
- name: source
dtype: string
- name: id
dtype: string
splits:
- name: val
num_bytes: 180989518
num_examples: 1000
download_size: 176065039
dataset_size: 180989518
- config_name: kashmiri
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
decode: false
- name: text
dtype: string
- name: source
dtype: string
- name: id
dtype: string
splits:
- name: val
num_bytes: 176844018
num_examples: 1000
download_size: 171833755
dataset_size: 176844018
- config_name: konkani
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
decode: false
- name: text
dtype: string
- name: source
dtype: string
- name: id
dtype: string
splits:
- name: val
num_bytes: 188111024
num_examples: 1000
download_size: 183620402
dataset_size: 188111024
- config_name: maithili
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
decode: false
- name: text
dtype: string
- name: source
dtype: string
- name: id
dtype: string
splits:
- name: val
num_bytes: 218230040
num_examples: 1000
download_size: 214156233
dataset_size: 218230040
- config_name: malayalam
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
decode: false
- name: text
dtype: string
- name: source
dtype: string
- name: id
dtype: string
splits:
- name: val
num_bytes: 166304661
num_examples: 1000
download_size: 159170882
dataset_size: 166304661
- config_name: manipuri
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
decode: false
- name: text
dtype: string
- name: source
dtype: string
- name: id
dtype: string
splits:
- name: val
num_bytes: 239607299
num_examples: 1000
download_size: 234255111
dataset_size: 239607299
- config_name: marathi
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
decode: false
- name: text
dtype: string
- name: source
dtype: string
- name: id
dtype: string
splits:
- name: val
num_bytes: 186090292
num_examples: 1000
download_size: 180083027
dataset_size: 186090292
- config_name: nepali
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
decode: false
- name: text
dtype: string
- name: source
dtype: string
- name: id
dtype: string
splits:
- name: val
num_bytes: 185127838
num_examples: 1000
download_size: 179502943
dataset_size: 185127838
- config_name: odia
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
decode: false
- name: text
dtype: string
- name: source
dtype: string
- name: id
dtype: string
splits:
- name: val
num_bytes: 142302383
num_examples: 1000
download_size: 137556904
dataset_size: 142302383
- config_name: punjabi
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
decode: false
- name: text
dtype: string
- name: source
dtype: string
- name: id
dtype: string
splits:
- name: val
num_bytes: 191888007
num_examples: 1000
download_size: 186733812
dataset_size: 191888007
- config_name: sanskrit
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
decode: false
- name: text
dtype: string
- name: source
dtype: string
- name: id
dtype: string
splits:
- name: val
num_bytes: 199595923
num_examples: 1000
download_size: 194073760
dataset_size: 199595923
- config_name: santali
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
decode: false
- name: text
dtype: string
- name: source
dtype: string
- name: id
dtype: string
splits:
- name: val
num_bytes: 192797505
num_examples: 1000
download_size: 173797216
dataset_size: 192797505
- config_name: sindhi
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
decode: false
- name: text
dtype: string
- name: source
dtype: string
- name: id
dtype: string
splits:
- name: val
num_bytes: 189076801
num_examples: 1000
download_size: 184160959
dataset_size: 189076801
am-pranav/stt-unified-bench
Private, language/locale-partitioned mini-benchmark for STT models.
Each subset is a dataset config (e.g., en, de, en_indian_accent, hi_in, en_us) with a single split val.
Audio is staged at 16 kHz and stored in-repo for reproducibility.
Schema
audio:Audio(sampling_rate=16000, decode=False)text: reference transcriptionlang: implied by dataset config namesource: upstream dataset tagid: source-stable id
⚠️ For internal evaluation only. Respect upstream licenses.