Datasets:

useKled
/

code-switched-writing

pretty_name: Code-Switched Messages
license: other
license_name: kled-commercial-training-license
license_link: https://www.kled.ai/datasets/code-switched-writing
language:
  - tl
  - en
  - id
  - ms
task_categories:
  - text-generation
  - text-classification
tags:
  - kled
  - commercial-license
  - consented-data
  - human-collected
  - sample-on-request
  - text
  - everyday-writing
  - phone-capture
  - southeast-asia
  - indonesia
  - philippines
  - malaysia
  - labeled
  - code
  - reasoning
  - code-switching
size_categories:
  - 10K<n<100K
annotations_creators:
  - crowdsourced
language_creators:
  - crowdsourced
source_datasets:
  - original
multilinguality:
  - multilingual
viewer: false
gated: manual
extra_gated_heading: Request a sample set of Code-Switched Messages
extra_gated_prompt: >-
  Kled delivers sample sets and licensed data by email, not through Hugging
  Face. Tell us a little about your work and the Kled data team will send a
  sample set and pricing. Your request goes to team@kled.ai.
extra_gated_fields:
  Company or lab: text
  Role: text
  What are you building: text
  Volume you need:
    type: select
    options:
      - Sample set only
      - Under 10,000 items
      - 10,000 to 100,000 items
      - 100,000 or more items
      - Full dataset
      - Custom collection on the Kled contributor base
  I agree to be contacted by Kled about this request: checkbox
extra_gated_button_content: Request a sample set
extra_gated_description: Prefer email? Write to team@kled.ai with the dataset name in the subject line.

Dataset Card for Code-Switched Messages

Code-Switched Messages

Informal messages written the way bilingual people text, switching languages mid-sentence, language pair declared.

Collected through the Kled app by consenting contributors and licensed for commercial training. The files are delivered by Kled under license, not downloaded from this repository. Use Request a sample set at the top of this page, or write to team@kled.ai. Catalog entry: kled.ai/datasets/code-switched-writing.

Dataset Details

Dataset Description

Contributors declare the two languages they mix, then write as they would message a friend who speaks both, switching inside the sentence, not between paragraphs. Monolingual text, sentence-boundary switching and anything translated are refused, and the brief tells people not to force it. Recruitment is confined to 11 Southeast Asian countries, where Taglish, Indonesian-English and Malay-English mixing is the ordinary way to text; three pieces per author at most, each tagged with its language pair. A large share of the world texts in two languages at once and almost none of it reaches training data, because it rarely appears in published text.

Curated by: Kled (useKled)
Collected through: Typed in the Kled app
Language(s): Filipino, English, Indonesian, Malay
Regions: Southeast Asia (11 countries)
Volume: 50,000+ pieces (approx. 250 MB)
Contributors: 25,000+
Modality: Text
Format: Plain text
Resolution: Plain text, UTF-8
Annotations: Labeled (declared language pair)
Quality control: 99%+ passed QA review
License: Non-exclusive commercial training license (terms); exclusivity available on request
Consent: Each contributor opted in to this use when they uploaded.

Dataset Sources

Catalog page: kled.ai/datasets/code-switched-writing
Sample set: through the access request at the top of this page, or team@kled.ai
Publisher: huggingface.co/useKled

Uses

Direct Use

Code-switched language modeling
Token-level language identification
Informal chat-style generation

Out-of-Scope Use

Redistribution or resale of the files, use outside the terms of the commercial training license, and any attempt to identify or contact contributors or people who appear in the data. Uses that require labels this set does not carry should be scoped with the data team first.

Dataset Structure

Data Instances

One piece per record, delivered as Plain text. Volume: 50,000+ pieces.

Data Fields

File: the Plain text file as captured or uploaded by the contributor
Labels: Labeled (declared language pair), entered by the contributor at capture and checked at review.
Capture metadata: capture tool, country of capture, submission time
Review status: QA outcome (99%+ passed QA review)

Field names and the manifest format are confirmed with the sample set.

Data Splits

Delivered as one set. Subsets by country, language or label, and custom splits, are available on request.

Dataset Creation

Curation Rationale

The dataset exists because the public alternatives named in the description are non-commercial, synthetic, small or collected elsewhere. Kled runs the task as a paid brief in its app so that the data is captured on purpose, under the capture rules above, by people who have opted in to commercial use.

Source Data

Data Collection and Processing

Contributors complete the task in the Kled app. Capture tool: Typed in the Kled app. Each submission is reviewed against the task brief; submissions that break the capture rules are rejected. 99%+ passed QA review. Files are delivered as Plain text, in the resolution noted above.

Who are the source data producers?

Kled contributors, who complete paid data tasks on their own phones. This task was open to contributors in the 11 Southeast Asian countries where the task runs. Kled has 625,000+ contributors in 100+ countries, strongest in Southeast Asia, the United States and Europe.

Annotations

Annotation process

Labeled (declared language pair). The label is entered by the contributor as part of the task and checked at review against the task brief; it is a contributor label, not an expert annotation, unless the description above says otherwise.

Who are the annotators?

The contributors themselves, at the moment of capture. Kled reviewers check each submission against the task brief.

Personal and Sensitive Information

Every item was contributed by its owner, who opted in to this use at upload. The task brief states what may and may not appear in a submission, and reviewers reject submissions that break those rules. Kled's data handling practices are documented at trust.kled.ai; questions about handling requirements for your use go to team@kled.ai.

Bias, Risks, and Limitations

Coverage follows the contributor base: strongest in Southeast Asia, the United States and Europe, thinner elsewhere, and limited to the 11 Southeast Asian countries where the task runs for this task. Devices, settings and conditions vary by contributor, which is the point of the collection and also means no two items are captured under identical conditions. Where labels are contributor-entered, they are checked at review but are not expert annotations. Volumes are the current scope of the task and grow as collection continues.

Recommendations

Request a sample set and run it against your task before licensing. Ask for per-country or per-language counts where they matter to your use.

Licensing Information

Non-exclusive commercial training license; exclusivity on request. Each contributor opted in to this use when they uploaded. The license terms are on the catalog page.

Citation

@misc{kled_code_switched_writing_2026,
  title        = {Code-Switched Messages},
  author       = {Kled},
  year         = {2026},
  url          = {https://www.kled.ai/datasets/code-switched-writing},
  note         = {Commercially licensed dataset collected through the Kled app. Sample sets on request.}
}

Dataset Card Contact

Kled data team: team@kled.ai