Chat, Audio and Short Film in One Evening: Categorizing AI Tool Use

One platform can contain several diary activities

A personal usage log becomes difficult to interpret when the platform name is used as the only activity category. A person might spend an evening reading text responses, listening to narrated audio and watching short videos within the same product. Counting the entire period as one type of AI interaction hides differences in what the participant actually did.

This is an increasingly relevant classification problem as digital products combine conversational interfaces with audio and video formats. The same account or character can appear across multiple media, while the participant moves between reading, listening and watching. A diary designed around older distinctions between chat applications and video platforms may struggle to describe such an evening accurately.

The solution begins with a distinction between the tool, the format and the task. Tool identifies the product or service. Format identifies the mode of interaction. Task describes the participant's activity or purpose. These dimensions can coexist without requiring a new category for every combination.

• Tool: the platform or application.
• Format: text, audio, short film or mixed media.
• Task: conversation, listening, viewing or another defined activity.
• Context: alone or with others.

Why chat, audio and film need separate labels

Text chat generally involves reading and entering written messages. Narrated audio involves listening, even when it belongs to a continuing conversational experience. Short film involves viewing moving images. These activities can share a broader purpose, but they place different demands on the participant's time and may occur in different circumstances.

For example, a participant could spend 12 minutes reading and responding to text, then listen to a narrated audio segment and later watch a short film. A diary that records only the platform might describe the entire sequence as conversational use. A diary with format labels can preserve the differences without creating three unrelated product categories.

The distinction also matters when activities overlap. Someone may listen to audio while preparing food or watch a short film while sitting with another person. The diary should record the primary activity and relevant context according to a stated rule. If simultaneous activities matter to the research question, an additional overlap field can be used.

A concrete example of a multi-format platform

TheFriend illustrates how a single AI companion product can bring several media formats into one environment: it offers text chat, multi-part voice confession stories and vertical short films of about one minute. The combination makes it possible for a diary entry to involve reading, listening and viewing while remaining associated with one platform.

For diary purposes, the important distinction is the activity taking place during each recorded interval. The product name identifies the tool, while the format label identifies whether the participant was using text, audio or film. This description keeps the classification neutral and avoids assuming that every format serves an identical purpose.

A practical method for recording an evening

Begin by recording the approximate start and end of the overall period. Then divide the evening into activities according to the recording rules. A format change is a useful boundary when the participant moves from reading to listening or viewing. A short interruption, however, does not necessarily require a new session if the underlying activity continues.

Consider an illustrative evening beginning at 7 p.m. A participant spends 15 minutes in text chat, listens to audio for 20 minutes, watches a one-minute film and then returns to text for another 10 minutes. The diary can preserve four activity intervals while associating them with one tool and one broader leisure period.

This example also demonstrates why total time and active format time should be reported separately. The platform-associated period is approximately 46 minutes if the intervals are consecutive. The format-specific totals are 25 minutes of text, 20 minutes of audio and one minute of film. The calculation describes the example's recorded intervals, not a typical usage pattern.

• Record the overall start and end time.
• Mark each meaningful format transition.
• Identify the activity associated with each interval.
• Record interruptions and overlapping activities consistently.
• Calculate format totals separately from the overall period.

Distinguishing a session from an activity

A session can be defined as one continuous period of engagement with a platform, while an activity is a particular action within that period. The distinction is useful when someone alternates between several formats without leaving the product. A 40-minute session could contain multiple shorter activities, each with its own format label.

A diary needs an explicit rule for interruptions. One practical approach is to treat a pause of five minutes or more as a separate interval when the participant has moved to another activity. This is a recording convention, not a universal definition of digital sessions. A study concerned with short interruptions might choose a different threshold.

Switching from text to audio and back to text can therefore remain one overall session while producing three activity records. Keeping both levels of information allows researchers to examine platform-associated time and the composition of that time without counting the same minutes twice.

What a personal usage log can reveal

A well-structured log can describe when multi-format use occurs, which formats occupy recorded intervals and how often a participant changes between them. Across several evenings, it may show that one format appears more frequently on certain days or that different formats are associated with different leisure contexts.

Such observations remain descriptive. Time spent listening does not establish how closely someone paid attention. Film duration does not establish how much the participant enjoyed it. A return to text does not automatically indicate dissatisfaction with audio or video. These interpretations require information beyond the activity record.

The diary also has to account for uncertainty in time estimates. A participant who records activities after an evening may remember the broad sequence but overlook brief interruptions. Recording start and end times during the activity can improve temporal detail, although it adds effort and can affect the ordinary experience being documented.

Keep categories stable as formats change

The strongest category system is one that can accommodate new formats without requiring a complete redesign. A small set of stable dimensions makes comparisons easier across products and recording periods. If a new platform adds narrated stories, interactive clips or another media type, the format field can expand while the task and context definitions remain consistent.

Category changes should be documented. If the definition of audio changes halfway through a diary project, earlier and later entries may no longer be directly comparable. A researcher should record when a category was introduced and decide whether previous entries can be classified using the revised rule.

The next step is to add separate tool, format, task and context fields to a personal usage log. Use them for the next seven evenings, record each format transition according to one stated session rule and calculate the total time for each format at the end of the recording period.

Questions readers ask