Subtitle meaning review: six recorded examples
The downloads contain six original inputs, their expected labels and one recorded API observation per input revision. Disagreements stay visible.
Provide paired English and Spanish cues with neighboring context and translation requirements. Review omissions, reversals or unsupported additions while allowing reordered wording that preserves the full meaning.
Subtitle cues divide a stream of speech, so a word absent from one target cue may appear in the next. Compare a complete neighboring window before deciding that meaning was lost.
This page uses original short English and Spanish lines. It checks meaning under the supplied requirements, including an explicit glossary; it does not grade timing, reading speed or visual synchronization.
The linked subtitle tool uses Jev to route potential translation defects for review. Translation generation is a separate step. This independent demonstration returns a review category and preserves both texts.
This is an independently authored text demonstration inspired by the linked X discussion. Its rules and synthetic inputs are ours. It does not reproduce the author’s application, verify reported outcomes, or perform an external action.

Use the sample to understand the rule, then test your own input for subtitle meaning review.

Provide paired English and Spanish cues with neighboring context and translation requirements.

Open the decision rules and keep the intended distinction between Preserved, Meaning defect, Review. Edit the choices only when testing your own rule.

Use a sample first, then try your own input. Compare the selected label with the other returned probabilities.

Review Review before using a result. The label is a decision about the supplied information, not an executed action.
| Original input | Observed label | Leading probability |
|---|---|---|
| Meaning across two cues | Preserved | 99.0% |
| Negation removed | Meaning defect | 99.0% |
| Number changed | Meaning defect | 99.0% |
| Authorized product name | Preserved | 98.0% |
| Incomplete neighboring context | Review | 84.0% |
| Unsupported language pair | Review | 100% |
These are six original synthetic inputs, tested once per request revision. The percentage is the leading option probability, not measured accuracy. Authored expectations and any disagreements remain in the downloadable evidence. 6 of 6 results matched the authored fixture expectations.Verified
The downloads contain six original inputs, their expected labels and one recorded API observation per input revision. Disagreements stay visible.
The single-request SDK runner uses your private TYPESAFE_API_KEY. The offline consumer reads saved observations, checks request hashes and prints a pending output. It does not call the model or change any external system.
# Offline: inspect the downloaded saved observations. No API key is needed.
python preview_choice_workflow.py samples.json saved-api-results.json
# Optional live request: install the SDK and set TYPESAFE_API_KEY privately.
npm install @typesafe-ai/sdk@0.6.0
node run-choice-sample.mjs samples.json case-1
# The live response prints to the terminal; it does not update saved-api-results.json.The input contract and these limits define what the result means.
Yes. The example compares meaning across the supplied window rather than demanding a word-for-word match within each cue.
No. Timing, reading speed, visual context and other language-quality checks remain separate. The label only addresses the supplied meaning comparison.
It describes the distribution over the supplied choices for this request. These six synthetic examples are functional tests with authored expectations, not a production accuracy benchmark.
No. The demo displays a label and distribution. The downloadable consumer produces a pending preview packet. Application code and a separate authorization policy would control any real action.
Edit the example above to see how a changed fact affects the decision, or open the Playground for another task.