AI is transforming the way we communicate across languages, making conversations that once seemed impossible, possible. For, a voice translation and dubbing AI platform, I had the chance to help make these conversations even smoother. I embarked on a journey to enhance user interactions, making the process more intuitive and enriching.


01/ Overview stands at the forefront of speech AI technology, offering hyper-realistic voice translation and dubbing across the broadest range of languages.

However, despite its innovative capabilities, the platform faces challenges in optimizing the user experience, particularly in areas crucial for the smooth creation and management of projects.

One such area, adding speaker details, is essential for producing authentic and contextually accurate dubbed content. This phase of the user experience was identified as having significant room for improvement, with user friction points hindering efficiency and overall satisfaction.

02/ my role

Product Designer - Research, Journey Mapping, UX Design, Interaction Design, Visual Design, Prototyping


24 hours

04/ Tools

Figma, Figjam

05/ Highlights

Effortless speaker setup

A clear path from the get-go ensures a smooth dubbing process ahead.


Precision at a click

Identifying speakers is hassle-free with audio and video clip previews.


Adaptive profiles

Swiftly assign or reuse speaker profiles, making each dub as authentic as the original.


Intuitive editing on the fly

Don't let editing slow you down. Tweak details seamlessly, keeping the focus on creativity.


06/ Uncovering Opportunities for Improvement

I initiated a detailed journey mapping to closely examine the's user experience. By simulating common user actions—creating projects, uploading files, and adding speaker details—I aimed to identify any friction points in the platform’s workflow.

The purpose was to pinpoint obstacles that could hinder user satisfaction and efficiency. Understanding these friction points was crucial for devising targeted enhancements to streamline the user journey on

Technical jargon and interface clarity


The use of technical terms and a lack of clear instructions during setup creates barriers to a smooth user experience. Demystifying these terms and improving clarity can make the platform more accessible to users of all backgrounds.

Challenges in speaker detail customization


The task of adding and customizing speaker details—critical for generating accurate dubs—proves to be non-intuitive. Users struggle to locate settings for speaker information and express uncertainty regarding the impact of these details on the final product.

Difficulty in editing and reviewing content


The difficulties encountered in editing—from dialogue adjustments to timestamp modifications—highlight a need for more intuitive editing tools. Streamlining these features will not only improve the accuracy of the content but also boost user confidence.

These insights not only highlight opportunities to elevate the user experience but also emphasize the potential impact on the quality and accessibility of dubbed content.

07/ Narrowing the Scope to Speaker Details

After a thorough analysis of the user journey on, it became evident that enhancing the process of adding and customizing speaker details warranted focused attention. This decision was driven by a few critical considerations:


08/ Problem Statement

How might we ensure that users find entering and customizing speaker details intuitive and straightforward?

09/ Brainstorming and ideating

Recognizing the critical role that speaker profiles play in the accuracy and authenticity of dubbed content, I explored and assessed various approaches that could make adding and customizing speaker details more intuitive and less cumbersome for users.



The aim was to create pathways that are both intuitive and efficient. This approach ensures that ideas are clearly communicated to stakeholders, guaranteeing comprehensive understanding and thorough coverage of every design aspect.


11/ Enhancing the dubbing process

This section unfolds the journey through iterative design solutions, each targeting key modules within the dubbing workflow. Through a cycle of evaluation, feedback, and refinement, I aimed to address and resolve the nuanced challenges users faced, from speaker identification to dialogue synchronization.

Dialogue association

The guided workflow aims to simplify the process of associating dialogues with the correct speakers by incorporating video playback directly within the dialogue editing interface. This feature assists users in accurately identifying speakers for each segment of the dialogue.

How might we create a workflow that enables users to accurately associate dialogues with their respective speakers through intuitive video playback integration?


The option featuring a video panel pop-up adjacent to the dialogue, was chosen for its stronger visual association between text and video content. Despite potential interface challenges, its direct connection aids in more accurate speaker identification, a critical factor in enhancing dubbing quality. The first option's static position and potential to obscure important interface element posed significant usability concerns.

Speaker details input form

Improving the speaker detail input process to accommodate the dynamic nature of dialogue assignments, ensuring that the AI's speaker identification enhances user corrections and inputs.

How might we design the speaker details input process to adaptively incorporate user corrections, ensuring the AI's speaker identification aligns with the actual dialogue distribution?


Allowing users to select previously entered speaker profiles for new dialogues marks a significant improvement in the platform's usability and accuracy. This change addresses the critical issue of redundant data entry and the potential for inaccuracies in speaker identification that arose from the platform's initial assumptions. It does not only improves efficiency for the users but also elevates the authenticity and reliability of the dubbed output.

12/ Streamlined and Intuitive Speaker Detail Management

In refining the platform's user experience, the final design represents a culmination of insights and iterations aimed at optimizing the speaker detail entry process. Here’s how the enhanced workflow brings clarity and efficiency to the forefront.

Set up your crew

Kick things off smoothly by setting up your speaker profiles first. This early setup ensures a distraction-free editing path later on.


Effortless speaker identification

Identifying speakers is now just a click away. Watch the corresponding video clip as you review their dialogue, all integrated into one seamless view.


Swift profile assignment

Assigning or editing speaker profiles is quick and hassle-free. Enter new details or pick from already created profiles—either way, it's streamlined for speed. And if you need to adjust anything, edits are straightforward.


Or, just skip to editing

Can't wait to get your hands on the editing? There's a shortcut for that. With the intuitive editing feature, you can leap over steps without missing a beat.



To say 'Hi.
If you want to hear about what color I'm planning to dye my hair next.
For an opportunity. I'm actively looking for a full-time role starting May 2023.

©️ 2023 Pareshi Rajveer