EasyOctopus's picture
Release German support experiment and held-out evaluation
524b0ab verified
|
Raw History Blame Contribute Delete
2.21 kB
metadata
language:
  - de
license: apache-2.0
task_categories:
  - token-classification
configs:
  - config_name: default
    default: true
    data_files:
      - split: train
        path: data/train.jsonl
      - split: validation
        path: data/validation.jsonl
      - split: test
        path: data/test.jsonl

German dictation punctuation synthetic dataset

Apache-2.0. Scenario families and their 2 phrasing templates were assigned to a split before rendering. Normalised exact text comparisons found zero overlaps between the new splits and the earlier datasets named in provenance.json.

Split Messages Scenarios Templates Annotations
train 560 10 20 868
validation 64 4 8 104
test 112 7 14 176

Each text is a lowercase dictation with word-final punctuation removed. Each gold item identifies a zero-based whitespace-delimited word index and one mark. The reference restores commas, periods, question marks and colons. Decimal separators within words stay intact. Unpunctuated headings are negative controls.

The training messages and newly generated test messages are fictional. Repeated slot draws within a template are correlated. Split families are disjoint but use the same authoring process. These results do not measure protection or dictation quality on customer messages.

Download and use

This dataset was created by Mahir Isikli (EasyOctopus). Its companion checkpoint is EasyOctopus/german-dictation-punctuation-model.

Install Hugging Face Datasets and load the public repository by ID.

from datasets import load_dataset

dataset = load_dataset("EasyOctopus/german-dictation-punctuation-data", "default")
message = dataset["test"][0]
print(message["text"], message["gold"])

The repository includes example.py --self-test to check annotation alignment, hashes and split separation. Run uv run --script generate.py --output new-splits from a downloaded copy to regenerate the 3 scenario splits. A different seed changes slot values while retaining the scenario assignments. LICENSE, NOTICE and provenance.json record source attribution.