Hugging Face's logo Hugging Face
  • Models
  • Datasets
  • Spaces
  • Buckets
  • Docs
  • Enterprise
  • Pricing
    • Website
      • Tasks
      • HuggingChat
      • Collections
      • Languages
      • Organizations
    • Community
      • Blog
      • Posts
      • Daily Papers
      • Hardware
      • Learn
      • Discord
      • Forum
      • GitHub
    • Solutions
      • Team & Enterprise
      • Hugging Face PRO
      • Enterprise Support
      • Inference Providers
      • Inference Endpoints
      • Storage Buckets

  • Log In
  • Sign Up

Datasets:
SinhaSpeech
/
sinhala-asr-data

Modalities:
Text
Formats:
parquet
Languages:
Sinhala
Size:
1M - 10M
Tags:
speech
asr
sinhala
whisper
Libraries:
Datasets
Dask
Polars
License:
Dataset card Data Studio Files Files and versions
xet
Community
1
sinhala-asr-data / data /stratified_v5
15.8 GB
Ctrl+K
Ctrl+K
  • 1 contributor
History: 1 commit
Yohan2003's picture
Yohan2003
Add stratified_v5: register-spelling and word-boundary/ZWJ conjunct normalization applied on top of stratified_v4 (407 register pairs, 60 boundary pairs; 13,120 of 154,828 rows changed). Text-only changes, audio bytes unchanged from v4.
e6c471e verified about 1 month ago
  • test.parquet
    1.63 GB
    xet
    Add stratified_v5: register-spelling and word-boundary/ZWJ conjunct normalization applied on top of stratified_v4 (407 register pairs, 60 boundary pairs; 13,120 of 154,828 rows changed). Text-only changes, audio bytes unchanged from v4. about 1 month ago
  • train.parquet
    12.5 GB
    xet
    Add stratified_v5: register-spelling and word-boundary/ZWJ conjunct normalization applied on top of stratified_v4 (407 register pairs, 60 boundary pairs; 13,120 of 154,828 rows changed). Text-only changes, audio bytes unchanged from v4. about 1 month ago
  • validation.parquet
    1.62 GB
    xet
    Add stratified_v5: register-spelling and word-boundary/ZWJ conjunct normalization applied on top of stratified_v4 (407 register pairs, 60 boundary pairs; 13,120 of 154,828 rows changed). Text-only changes, audio bytes unchanged from v4. about 1 month ago