0 views
How to Set Up a Content Source to Achieve the Desired Tone of Video
We explain how the choice of source and project settings in Telematic determine the tone of the video — and what to do if the result does not sound as needed.
We explain how the choice of source and project settings in Telematic determine the tone of the video — and what to do if the result does not sound as needed.
Once during a platform demonstration, a client showed us a link to an industry article and said, “I want the video to sound lively, not like a summary of a manual.” The material was excellent — rich in details, with numbers and examples — but the draft indeed turned out to be rather dry. The reason was not the article itself or the quality of the adaptation, but the fact that the tone of the video in Telematic is determined not by the source itself, but by the combination of the source and the project settings surrounding it.
This distinction is critical for the outcome: two different sources with the same project settings will yield videos with a similar character, while the same source with different settings will produce completely different clips. Let’s explore how these two aspects work together and what to adjust to ensure the tone meets expectations from the very first attempt.
First and foremost, it is essential to remember: language and tone reside in the project settings, not tied to specific material. This means that if today’s video turned out too formal, the issue is almost never with the article or the transcription — it lies in how the project is set up as a whole. That’s why the first thing to check when dissatisfied with the result is not the source, but the project settings.
Although the tone is set at the project level, the type of source determines the raw material that the adaptation works with, which also affects the final feel of the video. RSS or industry feeds usually bring in a variety of styles — it’s easier to compile a neutral digest from them than to create a cohesive authorial story. A specific website provides a more consistent flow of material in style if the source platform already has its recognizable voice. A link to a specific article is the most predictable option: you know exactly what will form the basis of the video. Transcribing someone else's YouTube video brings a conversational, informal tone — a good foundation for a lively, less "bookish" tone. And manual text offers maximum control: you decide what will be included in the material and in what order.
If a conversational, "human" tone is needed for the video, transcribing someone else's video or using manual text is usually closer to the goal than a dry news feed.
It’s important to understand what artificial intelligence does during the adaptation stage: it does not invent new facts or discard details from the original material but adapts the style of presentation and volume to fit the tone and project settings. If the article had five arguments and three figures, they will remain in the adapted text — what changes is how they sound: drier or livelier, shorter or more detailed.
Therefore, if you are not satisfied with the tone of the video, the issue is usually not that "the AI made something up," but that the tone settings of the project do not match the expected result — and this can be corrected with settings, not by changing the source.
After the text has been adapted, two more tools come into play that significantly affect the perception of tone: the “script volume” slider and the narrative pace switch — dynamic, standard, or calm. For news or conversational material, a dynamic pace and a script mode closer to hybrid usually work better — this creates the feeling of a lively narrative. For expert, detailed content, a calm pace and a full text mode that retains the source's detail are often more suitable.
It’s also worth considering the chosen script mode: full text sounds detailed and thorough, hybrid — more compact and dynamic, while teaser — a short vertical announcement without a call to action, which inherently has a more "advertising" tone.
If after adaptation the result still does not meet expectations, the order of actions is as follows: first, check the tone and language in the project settings, then the script mode and narrative pace, and only as a last resort change the source material itself. In practice, the solution is almost always found in the first two steps.
It’s also important to keep in mind the thematic content filter and auto-analysis thresholds — they filter out blatantly irrelevant material even before adaptation, so the final tone is not "blurred" by unrelated topics accidentally included in the source.
When we explained this logic to the client at the beginning of the article, and he adjusted the project tone rather than the source, the second draft of the same material sounded exactly as he had envisioned from the very beginning — even though the original article remained the same.
How to Choose a Voice for Your Brand: Clone, Pre-Recorded Voice, or Your Own Recording
Setting Up Autopilot from Scratch: A Guide for the First Channel