Google Translate Conversation Improvement Idea.

Christian Enema Jeremiah
8 min readDec 21, 2023

--

Google , Google Design, Google Play Apps & Games Team

Behance Presentation Link

A large number of individuals, tourists, groups and students travel annually for vacation, sight-seeing, studies, etc. And these individuals interact with strangers, temporary friends, and locals in the environment.

During my travels in Europe, I came to realize that a significant number of these individuals utilize Google Translate as a means of communicating with those who speak languages different from their own. However, it has become apparent that this approach often presents inconveniences, particularly in situations where the translation process necessitates the sharing of a single device between conversing parties.

Two users conversing using Google Translate with one device. Image credit : furkanfdemir

This circumstance subsequently diminishes the overall efficiency and effectiveness of the translation endeavor. Moreover, an additional limitation I identified pertains to the absence of a feature within the application that would allow for the retention of conversation histories with specific users. This functionality, which I often find myself desiring when interacting with the elderly lady at the reception, would undoubtedly enhance the utility and practicality of the translation tool. Hence, I sort to help improve the experience for these group of users while using Google Translate.

1. Understand

  • What is the problem?
  • Problem Definition: The reliance on a single mobile device for language translation among multiple individuals (Friends or Strangers) poses significant limitations, hindering effective communication and collaboration across language barriers.
  • Scope: Reduce limitations of using a single mobile device for language translation among individuals, strangers or friends, while using Google Translate.
  • Who does this problem affect? Group(s) or individuals who experience similar problem.
    Tourists, international students, business people, etc. Individuals, friends, strangers who speak different languages but wants to communicate.
  • What is the size and impact of the problem? How big is the problem? what happens if this problem is not solved?
    Google translate has over 1billion users. most of which experience this problem while interacting. Link
  • Does solving this problem align to business goals and objectives? How does it align.
    Google Translate business goals is to connect people, places, and cultures without language barriers. Aims to break down language barriers and enable people from different cultures and backgrounds to communicate more effectively. Link

2. Research

Primary research: was conducted to observe how users interact using “Conversation” feature in a natural scenerio. Participants were recruited in groceries shops, public transports, and schools. The observation was achieved by introduction (using Google Translate), asking for help (Tasks), and lastly inquiries about the experience:

  • How often do you use conversation feature in google translate?
  • How comfortable are you while using conversation feature in google translate?
  • How do you feel about sharing a single device for translation when communicating with others?

8 users were observed which took an average of 3 - 4 mins for each participant.

Primary Research : Participant Responses

Secondary Research: Using Octoparse review data was meticulously collected from Google Play Store which comprising of over 3,759 Google Translate reviews spanning the years 2018 to 2023. Notably, a heightened emphasis was placed on the most recent year, 2023 (926 user reviews). Aimed towards gaining a better understanding of the problem. Subsequent to the data collection process, refinement was undertaken using Microsoft Excel, followed by a thorough analysis utilizing Python. The primary objective of this analytical was to further understand user pain-points related to the problem.

Secondary Research : Google Translate Android Mobile Rating Analysis

Step 1: Data Cleaning and Preparation: Before diving into analysis, the data was cleaned and properly formatted. Removing duplicates, missing values, and outliers. Converted date-related columns into a single date format for ease of analysis.

Step 2: Descriptive Analysis: The data analysis started with basic descriptive statistics to get an overview of the data.

  • Summary statistics for “Star ratings” to understand the average rating, standard deviation, and distribution. Average rating : 2.706, STD: 1.62 (the low standard deviation indicates the ratings are clustered and not scattered.)
  • The Image below shows frequency distribution of user ratings to identify the frequency of different rating degrees. The graph shows that the most common star rating in 2023 was “1” which shows users dissatisfaction in experience while using the Google Translate mobile app. But the graph does not explain the reason why. To further understand this, text analysis was conducted.
Google translate star rating frequency distribution for 2023 android users

Step 3: Text Analytics: To further understand the problem, the data was analyzed to discover frequent word occurrences. This is to help uncover clusters of comments with similar issues and words often associated with the focus problem (Improving google translate conversation).

The word cloud showed varieties of words related to translation, including Translation, Language, Text, Speech, Translate, Translations, Translation device, Translation app, Machine translation, Human translation

Overall, the word cloud suggests that there is a high demand for translation services and are looking for translation solutions that are easy to use and reliable. However Results from word cloud analysis did not indicate clusters of keywords related to the problem, which will require further testing after the prototype has been designed.

Google Translate Word Cloud Analysis

3. Design :

Based on the initial problem and research results, the design phase started off with identifying key personas, creating user journey map, user flow, ideation, and Hi-Fi prototype to show possible solutions on how this problem can be solved.

USER PERSONAS :

Adrian Boris

Demographics:

  • Age: 25–35
  • Gender: Male
  • Location: England
  • Occupation: Therapist

Psychographics:

  • Interests: Travel, exploring new cultures, meeting new people
  • Values: Adventure, authenticity, cultural experiences
  • Goals: To have an immersive and authentic travel experience, to learn about French culture, to make new friends

Gulnar Jerry

Demographics:

  • Age: 20
  • Gender: Female
  • Nationality: Azerbaijani
  • Location: Poland (Erasmus student)
  • Occupation: Student

Psychographics:

  • Goals: To learn about Polish culture, to make new friends, and to succeed in her studies.
  • Challenges: Communicating with Polish people, finding her way around, and dealing with health issues.
  • Interests: Travel, meeting new people, learning about different cultures.

USER JOURNEY MAP :

User journey map for Gulnar persona.

USER FLOW:

User flow for multi mode conversation

High Fidelity Prototype :

Google Translate Multi Mode Conversation

Mono Mode :

Google Translate’s Mono Mode is a basic mode of communication that allows two users to converse using a single mobile device. This mode is suitable for short conversations or for testing the translation feature. However, it is not the most efficient means of communication as it requires users to share the same device, which can be cumbersome and limiting. Additionally, the translation accuracy may be less reliable in this mode due to the shared device.

Google Translate Mono Mode

Duo Mode :

Google Translate’s Duo Mode is an improved mode of communication that allows two users to converse using their individual mobile devices. This mode is more efficient than Mono Mode as it enables users to have a more natural and uninterrupted conversation. Additionally, Duo Mode protects user privacy by not saving conversations, ensuring that personal information is not shared. This makes Duo Mode ideal for conversations with strangers or for situations where maintaining privacy is important.

Google Translate Duo Mode

Friend Mode :

Google Translate’s Friends Mode is specifically designed for individuals who are gradually becoming friends but speak different languages. This mode offers the same benefits as Duo Mode, including efficient communication and enhanced privacy. However, Friends Mode also encourages a more personal and meaningful connection by emphasizing the establishment of friendship bonds. This mode encourages users to share their thoughts and experiences beyond the mere exchange of information, bridging language barriers and fostering meaningful relationships.

Google Translate Friend Mode

In summary, Google Translate’s three communication modes cater to different needs and preferences. Mono Mode provides a basic translation tool for quick interactions, while Duo Mode enhances efficiency and privacy for extended conversations. Friends Mode goes a step further by promoting personal connection and fostering friendships between people from different language backgrounds. These modes demonstrate Google Translate’s commitment to breaking down language barriers and facilitating meaningful human interactions across cultures.

Benefits of Duo and Friend Mode to Google Translate :

  • Improved communication efficiency: Duo and Friend Mode allows users to use their individual devices to translate conversations, eliminating the need for a shared device. This makes communication more natural and and increases translation efficiency, especially for longer conversations.
  • Enhanced privacy: Duo Mode does not save conversations, ensuring that sensitive information is not shared. This is especially important for conversations with strangers or in situations where privacy is paramount.
  • Fosters meaningful connections: Friends Mode encourages deeper connections and personal interactions between users. It encourages users to share their thoughts and experiences beyond the exchange of information, helping to bridge language barriers and foster real friendships.
  • Expands user base: Duo and Friend Mode can attract a wider range of users, including those who need to communicate with multiple people or those who want to connect with friends from different cultures.
  • Strengthens Google Translate’s brand: Duo and Friend Mode demonstrate Google Translate’s commitment to innovation and its dedication to breaking down language barriers and promoting global communication. This can enhance the company’s reputation and attract more users.

Overall, Duo and Friend Mode offer significant benefits to Google Translate, including improved communication efficiency, enhanced privacy, and the potential to foster meaningful connections and expand the user base. These improvements align with Google Translate’s mission of connecting people without language limitations and can make a positive impact on global communication.

4. Next Steps:

Due to the limited time

Test:

  • Conduct user testing to gather feedback on the three communication modes.
  • Evaluate the effectiveness of each mode in facilitating language-independent communication.
  • Identify areas where the modes can be improved in terms of usability, efficiency, and privacy.

Iterate:

  • Based on user feedback, refine the features and functionalities of each mode.
  • Implement changes to enhance the overall user experience and address any identified issues.
  • Continuously improve the translation accuracy and reliability across all modes.

Improve:

  • Explore innovative ways to facilitate meaningful human connections and break down language barriers.

--

--

Christian Enema Jeremiah
Christian Enema Jeremiah

Written by Christian Enema Jeremiah

I am committed to contributing my skills and expertise to drive business success and create delightful user experiences through design innovation.

No responses yet