Arabic Perception Lab

Research platform · Arabicling

About Arabic Perception Lab

Design, pilot, and run browser-based experiments on how listeners perceive speech. Present your recordings, collect listener judgments, and export responses with their trial order and experimental conditions.

Request a researcher account Read the methods
Cite the lab Get the software reference for your article or thesis.

What you can study

Run multiple-choice identification tasks or collect open-ended responses in an Arabic or English interface. Applications include identifying dialects or speech sounds, judging prominence or sentence type, and evaluating how listeners perceive synthesized speech.

For example, prepare a set of vowel tokens that vary in duration, present one token per trial, and ask listeners to select the category they heard. Alfaifi’s study illustrates this design with Najdi and Cairene Arabic. Alfaifi (2026).

Prepare and measure the stimuli in your preferred acoustic software, then link the final files in the lab. Keep the stimulus files and record the speakers, recording conditions, and any manipulations of F0, duration, or intensity.

Published studies on Arabic prosody

These studies illustrate how acoustic analysis, listening experiments, and computational modelling can inform the design of a speech perception study.

Research methods and experimental controls

Listening conditions and headphone screening

The built-in headphone screen adapts the six-trial stereo phase-cancellation procedure described by Woods and colleagues. In each trial the listener chooses the quietest of three tones. One louder tone has opposite phase in the two stereo channels. The default passing score is five out of six on one attempt. The lab gives written volume instructions, and researchers can configure the threshold and retries or use their own listening check. Woods et al. (2017).

Treat the screening outcome as a quality-control measure. For studies that depend on calibrated sound levels or a controlled listening environment, verify the playback equipment and conditions separately.

Trial order, condition assignment, and playback

Choose fixed order, seeded randomization, or cyclic counterbalancing. Configure blocks, participant lists, practice trials, playback limits, and device requirements as needed. The participant export records the assigned list, seed, algorithm, and actual presentation order. Check achieved group balance after exclusions and attrition.

Response timing and data quality

When a timing reference is available, the lab records browser-based intervals to the first response, last edit, and submission. Choose the browser event for playback start, playback end, or response availability as the reference. Devices, browsers, and audio hardware can introduce latency. If small timing differences matter, measure latency on representative setups and document the precision you achieved. Bridges et al. (2020).

The studies cited here describe the methods or illustrate their use. Document how your particular experiment performs in a pilot with representative browsers, devices, and listening conditions before you recruit. Eerola et al. (2021).

Published methodological references

What researchers can document for reviewers

Each published experiment version retains its consent text, trial configuration, and stimulus references. Pilot responses are exported separately. The export files together include participant and stimulus identifiers, conditions, actual trial order, answers, timing fields, and available screening and playback diagnostics. These records let your team trace a result back to the protocol and stimuli used to collect it.

  • Describe participant eligibility, recruitment, consent, and any required institutional ethics approval. Agree with your institution on storage location, retention, withdrawal, and sharing. Participants’ browsers contact the providers that host externally linked audio.
  • Report stimulus preparation, task and response options, order and list assignment, playback rules, screening thresholds, and the timing reference if you analyze response times.
  • Record the pilot setup, prespecified exclusions, experiment version, and data collection dates. Share the materials, analysis code, and de-identified exports that your consent and approvals permit.
Example wording for a Methods section

We administered [task] using Arabic Perception Lab (Alzaidi, 2026). Participants heard [stimuli] in [presentation order and assignment]. We used [device, playback, and screening settings] and recorded [responses and timing reference, if applicable]. We piloted [browsers, devices, and listening conditions] and applied the prespecified [exclusion rules]. We analyzed data from experiment version [number] using [analysis approach].

How to cite the lab

Cite the lab as research software and cite the published methods you used separately. Record the software build during data collection and report it alongside your experiment version. They identify different parts of the study. Smith et al. (2016).

Alzaidi, M. S. A. (2026). Arabic Perception Lab [Computer software]. Arabicling. https://perception.arabicling.com/about

Current software build 39960543bee56d0f75bb3773701bd043e72c1800