This pilot test plan provides a framework for evaluating KUDO AI in multilingual classroom training environments. The objective is to determine whether instructors and learners can effectively communicate and participate in training sessions when participants speak different languages.
Pilot Objectives
This pilot is designed to evaluate how KUDO AI supports multilingual classroom training environments by increasing accessibility for learners who prefer training in their native language, reducing communication barriers during instructor-led sessions, supporting multilingual participation and classroom discussions, and enabling instructors to teach naturally without significant workflow changes.
The pilot is considered successful if participants can independently join the session using their personal devices, access translated audio and captions in their preferred language, follow training content and classroom discussions with minimal confusion, ask and answer questions during the session, demonstrate understanding of the material being presented, and express confidence in using the solution during future training sessions.
Before the Pilot
Before conducting the pilot, organizations should provide any available presentation materials, training agendas, assessment scripts, evaluation forms, acronyms, equipment names, procedures, and company-specific terminology, including Do Not Translate (DNT) terms and preferred glossary translations. Providing these materials in advance allows KUDO to customize terminology and optimize translation performance, resulting in a more accurate and representative testing experience.
Pilot Workflow
Display the Mobile Link QR code or share the Mobile Link with participants.
Participants join and select their preferred language.
The instructor delivers training content in the source language.
Participants receive translated audio and captions on their devices.
Learners ask questions and participate in discussions.
The instructor continues teaching using their normal speaking patterns.
Real-World Validation Questions
During the pilot, evaluate the following:
Can participants join the session without assistance?
Is translated content easy to understand?
Do participants prefer translated audio, captions, or both?
Can instructors teach naturally without significantly changing their delivery style?
Does the workflow support classroom discussions and learner engagement?
Would instructors and learners feel comfortable using this workflow in future training sessions?
Feedback Questionnaire
For Learners
How easy was it to join the session?
How easy was it to select your preferred language?
How easy was it to understand the translated content?
Did you prefer translated audio, captions, or both?
Did you miss any important information during the session?
Did translation help you participate more confidently?
Would you use this solution again?
For Instructors
Was it easy to deliver training while participants used translation?
Did you need to change your speaking style?
Were participant interactions easy to manage?
Would you feel comfortable teaching an entire class using this workflow?
Would you use this solution for future multilingual training sessions?
After the Pilot
At the conclusion of the pilot, evaluate translation accuracy, participant comprehension, ease of deployment, instructor effectiveness, classroom interaction and engagement, participant confidence and trust in the experience, and the overall readiness of the solution for multilingual training applications.
Our AI Translation system offers several advanced features tailored specifically for meetings with AI Speech & Captions. These advanced functionalities are designed to improve translation quality, ensuring accuracy, and maintaining consistency.
Glossary
Do-not-translate
Word Priority
AI Captions-only supports Word Priority only; the other features are not available in this modality.
At KUDO, we will take care to customize your personal translation engine with your company data. Our team will guide you on the best practices to collect the relevant information and will make sure that your AI engine is using your terminology. Please contact your KUDO point of contact to get in touch with our linguistic expert. Below, you’ll find a detailed overview of how each linguistic feature can benefit your translation process.
1. Glossary
The Glossary allows to define specific translations for terms in the listeners’ languages, ensuring consistency and precision. This feature supports different variations of the same word. Therefore, there is no need to enter plural forms or inflected verbal forms. Similarly, this feature is not case sensitive, which means that capitalization is not important. See example below:
Original sentence: “Hoy vamos a hablar sobre la AV que es un tema muy relevante.”
✔ with glossary ✗ without glossary
In a corporate setting, an acronym like "AV" in Spanish may have a default translation in English as "VA", but by using a glossary, it can be translated more accurately into its complete version "Visual Acuity" to reflect the complete meaning in context.
Max. size of glossary:
Each account can have up to 5 glossaries, with a maximum of 1,000 entries per glossary.
Where to access this feature:
Under the AI Customization tab, you can select or manage your glossaries.
Choose a glossary for each session in the Select Glossary panel.
Click Manage Glossaries to upload new ones or set a default glossary.
Each account supports up to 5 glossaries, with 1,000entries per glossary.
The active glossary is marked “Use for this meeting.”
EXPECTED FORMAT
Upload an XLSX file. Download the template and fill it in, using the English name of each language as the column heading (e.g. Spanish, French).
Commas are not allowed.
Do not include copyright or trademark symbols — the AI voice will read them aloud.
Enter only one term per cell; do not offer alternatives separated by a slash (e.g. VA/Visual Acuity).
Keep entries short. Single words, compound terms and brief phrases work best; full sentences may not be captured correctly by the AI.
Cells may be left empty for any language where the client has no preferred translation.
Supported Languages:
The Glossary feature is available in the following 32 languages:
The Do-Not-Translate (DNT) feature ensures that specific words or terms are not translated. This is especially helpful for preserving product names, proper nouns, brand names, and acronyms in their original form. Adding the speakers’ names to the DNT list is also recommended, as it prevents the name from being translated and improves its recognition during speech processing. This feature is case sensitive. Refer to the example below:
Original sentence: “Remember to have at your disposal the office package.”
✔ with DNT ✗ without DNT
For example, the product name "Office Package" could be incorrectly translated as “paquete de oficina” into Spanish, which loses the specific reference to Microsoft’s product suite. However, by adding the word "Office" to the DNT list, we ensure that it remains untranslated, resulting in the correct translation of “paquete Office.” This maintains brand consistency and preserves the intended meaning.
Max size of DNT:
The number of words that can be added to the DNT list is limited to 500 words.
Where to access this feature:
The DNT feature can be edited autonomously under the "AI Customization" tab. The list of terms added can be saved for upcoming meetings by clicking on the bottom option “Use these terms automatically in future meetings”.
Expected format:
Each entry can be a single word or a short phrase, including compound terms, and entries must be separated by commas. The AI looks for an exact match: if "Fardad Zabetian" is added to the list and the speaker says only "Fardad", the term will not be recognized, because the AI needs the surname to complete the match. For this reason, we recommend adding first name and surname as two separate entries, e.g. "Fardad", "Zabetian".
Supported Languages:
This feature is available in all speaking languages.
Recommendations:
When adding acronyms, avoid including periods between letters (e.g., "ABC" instead of "A.B.C.").
Note that this feature does not account for variations or extensions of the word. For example, if you add "car" to the system, related forms such as "cars" will not be automatically recognized. Ensure all necessary variations are explicitly added.
Do not add common nouns or words that carry another meaning in a different context — e.g. "Will". If added to the DNT list, "will" would never be translated, including when it is used as the future-tense auxiliary verb (as in "I will go"), because the AI has no way to distinguish the name from the grammatical usage. Adding terms like this can have a significant negative impact on translation quality, so use extra caution before adding any common word.
3. Word Priority
Word Priority (WP) is a feature that ensures key terms are accurately recognized; whether they should be translated or kept in their original form. It is supported across all speaker languages. Word Priority is the only customization feature that works in both modalities.
Max size of DNT:
The number of words that can be added to the WP list is limited to 500 words.
Where to access this feature:
The WP feature can be edited autonomously under the "AI Customization tab". The list of terms added can be saved for upcoming meetings by clicking on the bottom option “Use these terms automatically in future meetings”.
Expected format:
Each entry can be a single word or a short phrase, including compound terms, and entries must be separated by commas. The AI looks for an exact match. For example, if I add “KUDO AI” to the list but the speaker only says “KUDO,” it may not be recognized because the AI is expecting the complete term “KUDO AI.” If both forms may be used during the meeting, add them as separate entries: “KUDO” and “KUDO AI.”
Supported Languages:
This feature is supported in all speaking languages.
Recommendations:
One-way communication (AI audio and captions) modality: Use Word Priority to ensure accurate recognition of technical terms when:
You don’t have the expected translation in a glossary, or
You want the term recognized correctly but still allow it to be translated.
Interactive communication (AI captions only) modality: Use Word Priority to ensure accurate recognition of:
Proper names
Brand or product names
Any other key term in the speaker’s language.
Conclusion
KUDO AI’s advanced features are integral to enhancing the overall user experience in real-time translation systems, providing a more tailored and accurate translation. These capabilities allow for greater customization, particularly in the recognition and translation and complex terms. This level of personalization not only boosts translation accuracy but also significantly improves communication in multilingual settings, leading to higher user satisfaction and a more seamless, professional experience.
This pilot test plan provides a framework for evaluating KUDO AI in one-on-one, multilingual networking and marketplace environments. The objective is to determine whether Subject Matter Experts (SMEs) and attendees can effectively communicate in real time when they speak different languages.
For the purposes of this evaluation, the sample use case is based on a tradeshow or networking booth where an English-speaking SME interacts with attendees who may speak one of several different languages. Using the KUDO Live application, SMEs can create multilingual meetings on demand, share a QR code with attendees, and immediately begin communicating through real-time AI translation.
Unlike classroom training scenarios, this pilot focuses on spontaneous, ad hoc conversations that occur in busy event environments where participants may have only a few minutes to interact. The goal is to determine whether KUDO can effectively support multilingual networking, product demonstrations, information sharing, and brief one-on-one discussions without requiring bilingual staff or professional interpreters.
Pilot Objectives
This pilot is designed to evaluate how KUDO AI supports multilingual one-on-one conversations by increasing accessibility for attendees who prefer to communicate in their native language, supporting spontaneous discussions in dynamic event environments, and enabling SMEs to engage naturally with multilingual attendees without requiring bilingual staff or professional interpreters.
The pilot is considered successful if participants can independently join the session using their personal devices, access translated audio and captions in their preferred language, communicate with minimal confusion, ask and answer questions naturally, and express confidence in using the solution during future networking and marketing interactions.
Before the Pilot
Before conducting the pilot, organizations should determine the languages that will be supported during the event and identify any company-specific terminology, product names, acronyms, or industry terminology that may be discussed during conversations. Providing these materials in advance allows KUDO to customize terminology and optimize translation performance, resulting in a more accurate and representative testing experience.
Pilot Workflow
The SME creates a meeting on demand using the KUDO Live application.
The meeting includes English and the required translation languages.
An attendee approaches the booth and requests a preferred language.
The SME shares the Viewer QR code with the attendee.
The attendee scans the QR code, selects their language, and immediately receives translated audio and captions on their device.
For two-way conversations, the attendee joins using the Participant link on their own device or a preconfigured KUDO device.
The SME and attendee communicate naturally through real-time translation.
Real-World Validation Questions
During the pilot, evaluate the following:
Can SMEs create meetings quickly and independently?
Can attendees join without assistance?
Is translated content easy to understand?
Do attendees prefer translated audio, captions, or both?
Can participants ask and answer questions naturally?
Does the workflow support spontaneous networking conversations?
Can SMEs communicate effectively without bilingual support?
Would participants feel comfortable using this workflow during future events?
Feedback Questionnaire
For Attendees
How easy was it to join the session?
How easy was it to select your preferred language?
How easy was it to understand the translated content?
Did you prefer translated audio, captions, or both?
Did you miss any important information during the conversation?
Did translation help you communicate more confidently?
Would you use this solution again?
For SMEs
Was it easy to create and manage the meeting?
Were attendees able to join without assistance?
Could you communicate naturally during the conversation?
Were participant questions easy to understand?
Would you feel comfortable using this workflow during future events?
Would you use this solution again?
After the Pilot
At the conclusion of the pilot, evaluate translation accuracy, ease of deployment, attendee comprehension, communication effectiveness, onboarding speed, participant confidence and trust in the experience, and the overall readiness of the solution for multilingual networking, tradeshow, marketplace, and one-on-one engagement applications.
The Meet Now feature allows you to create an instant multilingual meeting directly from the KUDO Live mobile application. Within seconds, you can launch a meeting, invite participants, and begin using AI-powered speech translation and multilingual captions.
Before You Start
For the best experience:
Download the KUDO Live mobile app on all participating devices.
Use earphones or headphones whenever possible to improve audio quality and reduce echo.
Ensure each device has a stable Wi-Fi or cellular connection.
Allow camera and microphone permissions when prompted.
Open the KUDO Live app and sign in using your KUDO account credentials.
Tap Meet Now to create a new multilingual meeting.
3. Configure Your Meeting
Enter the meeting details then tap Start.
Important: Speaker and Translation Languages cannot be modified after the meeting begins.
4. Join as a Participant
Participants can join the meeting and actively speak using their device microphone. This enables two-way multilingual conversations with real-time AI translation.
Invite Participants
5. Share the Meeting QR Code
Tap the Share icon within the meeting to display the participant QR code.
Participants can scan the QR code to join instantly from their own mobile devices.
Tips for Best Results
Speak clearly and at a natural pace.
Use headphones whenever possible to avoid audio feedback.
Avoid multiple people speaking at the same time.
Keep your device connected to a stable internet connection.
Ensure all required languages are configured before starting the meeting.
Consider using Captions Only meeting mode for rapid 1:1 conversations.
Troubleshooting
Symptom
Possible Cause
Resolution
Participants cannot hear translated audio
Incorrect listening language, low volume, or audio device issue
Verify the selected listening language, increase device volume, and confirm headphones or speakers are connected.
Speech is not translating correctly
Incorrect speaking language or poor audio quality
Verify the selected speaking language, move closer to the microphone, and reduce background noise.
Participants cannot join using the QR code
Meeting inactive, outdated application, or connectivity issue
Confirm the meeting is active, update KUDO Live to the latest version, and verify internet connectivity.
Create an AI session with the languages, Host, and attendee access your meeting needs.
Before you start
Have your KUDO Client credentials and domain, audience size, spoken languages, translation languages, and terminology ready.
Steps
Sign in to your KUDO Client account. Select New Meeting, then AI Translation.
Choose speech with captions for spoken translation, or captions only for text-focused discussions.
Set speaker languages for incoming speech and translation languages for attendee output.
Assign the Host and prepare terminology through AI Customization.
Distribute the appropriate Participant, Operator, Viewer, or Mobile Link. Use the Widget when embedding access in a website.
Rehearse joining and language selection before the session.
What to expect
The Host manages meeting controls; attendees receive access through the links selected for their roles.
Changing a live meeting
Use Edit Meeting to change translation languages and Update to apply them. Speaker languages and AI Customization remain unchanged during the live session.
Join with your Participant/Speaker link and choose your caption and listening languages. Use a headset before starting.
Audience: Hosts, Operators, and Presenters.
Choose your access method
Follow the desktop or KUDO Live steps below for Participant/Speaker access.
If your organizer supplied a Mobile Link or QR code for translated speech and captions, open it in a supported smartphone or desktop browser. That access method requires no login or software download.
Enable automatic detection so KUDO can identify incoming speech among the languages configured for your session.
Before you start
Decide which languages speakers will use. Detection supports up to ten configured speaking languages and must be enabled during scheduling.
Steps
Configure the expected speaking languages while scheduling the meeting. Enable automatic language detection before it starts.
Where the participant’s speaking-language menu is available, choose Automatic.
Rehearse language changes using clear, steady speech, one speaker at a time.
What to expect
Only configured languages are detected. Switching can take several seconds and multiple phrases to stabilize; it is not instantaneous. Avoid changing language mid-sentence or speaking over others.
Troubleshooting
Issue / Task
Possible Cause
Recommended Action
Detection option is unavailable
Not enabled during scheduling
Create a new session with detection enabled.
Wrong language is recognized
Speaking-language setting mismatch
Choose Automatic or select the correct language manually.
Output slows after switching
Rapid changes or overlapping speech
Pause between languages and let one person speak at a time.
KUDO includes AI Customization tools designed to improve translation accuracy, terminology consistency, and overall multilingual meeting quality. These features help control how specific words, names, acronyms, and technical terminology are recognized, translated, or preserved during AI processing. Overall AI translation quality also depends heavily on microphone quality, stable internet connectivity, and clear speech from participants throughout the meeting.
Audience: Hosts, Operators, Presenters, and Meeting Organizers.
Before You Start
Before uploading a glossary, confirm that you have access to a KUDO Client account, the glossary file is prepared in the supported format, terminology has been reviewed for spelling and language consistency, and the glossary languages match the languages configured for the meeting.
Keep in mind that glossaries improve translation consistency but may not override every contextual translation generated by the AI engine. For terms, names, or acronyms that should never be translated, use Do Not Translate (DNT) alongside the glossary.
Uploading a Glossary
Go to live.kudoway.com and sign in with your KUDO Client credentials.
Click your Account Profile in the top-right corner of the dashboard.
Select My Glossaries from the dropdown menu.
Click Upload Glossary, select the file from your computer, and complete the upload. Once uploaded, the glossary will be available during AI Meeting setup and configuration.
Troubleshooting
Issue
Action
Expected Result
Terms not translating consistently
Review spelling, language mapping, and duplicate entries
Translation consistency improves for configured terminology
Proper names or acronyms still being translated
Add the term to Do Not Translate (DNT)
The term remains unchanged in translated output
Glossary entry is being ignored
Verify the term exactly matches what is spoken
The glossary entry applies correctly during translation
KUDO includes AI Customization tools designed to improve translation accuracy, terminology consistency, and overall multilingual meeting quality. These features help control how specific words, names, acronyms, and technical terminology are recognized, translated, or preserved during AI processing. Overall AI translation quality also depends heavily on microphone quality, stable internet connectivity, and clear speech from participants throughout the meeting.
Audience: Hosts, Operators, Presenters, and Meeting Organizers.
Do Not Translate (DNT)
The Do Not Translate (DNT) feature preserves specific words or terms in their original language instead of translating them during AI processing. It is commonly used for speaker names, acronyms, product names, company names, and other proper nouns that should remain unchanged across multilingual meetings. For example, adding “Office” to Do Not Translate prevents the term from being translated into another language during AI translation processing.
The Do Not Translate (DNT) list supports up to 500 words per account configuration. The feature is available directly from the KUDO meeting scheduling workflow under AI Customization > Do Not Translate, where organizers can create and manage terms that should remain unchanged during AI translation processing.
Supported Languages
The DNT feature is available in all 49 supported speaker languages.
Arabic
Basque
Bulgarian
Cantonese
Catalan
Chinese
Croatian
Czech
Danish
Dutch
English (US, UK)
Estonian
Filipino
Finnish
French (France, Canada)
Galician
German
Greek
Hebrew
Hindi
Hungarian
Indonesian
Irish
Italian
Japanese
Khmer
Korean
Latvian
Lithuanian
Malay
Maltese
Marathi
Mongolian
Norwegian
Persian (Farsi)
Polish
Portuguese (Portugal, Brazil)
Romanian
Russian
Slovak
Slovenian
Spanish (Spain, LatAm)
Swedish
Tamil
Thai
Turkish
Ukrainian
Vietnamese
Welsh
Troubleshooting
Issue / Task
Possible Cause
Recommended Action
Glossary terms are not translating as expected
Term format does not match the spoken term
Verify the Glossary entry matches the exact term expected to be spoken during the meeting
Do Not Translate terms are still being translated
Alternate spelling or variation not added
Add additional variations, plurals, or acronyms as separate DNT entries
AI translations appear inconsistent across languages
Multiple translation preferences configured for the same term
Use only one preferred translation per Glossary entry
Acronyms are not recognized correctly
Acronyms include punctuation
Add acronyms without periods or special characters (example: “ABC” instead of “A.B.C.”)
AI customization changes are not applied
Meeting already started
AI Customization settings should be configured before the meeting begins
KUDO includes AI Customization tools designed to improve translation accuracy, terminology consistency, and overall multilingual meeting quality. These features help control how specific words, names, acronyms, and technical terminology are recognized, translated, or preserved during AI processing. Overall AI translation quality also depends heavily on microphone quality, stable internet connectivity, and clear speech from participants throughout the meeting.
Audience: Hosts, Operators, Presenters, and Meeting Organizers.
Word Priority
The Word Priority feature improves recognition accuracy for important terms in the speaker’s language during AI processing. It is commonly used to help the AI better recognize technical terminology, proper names, brand names, product names, and multilingual event terminology without forcing a fixed glossary translation. Word Priority is best used when a term should be recognized correctly by the AI, while still allowing the translation engine to translate the term naturally within the context of the conversation.
The Word Priority list supports up to 500 words per account configuration. The feature is available directly from the KUDO meeting scheduling workflow under AI Customization > Word Priority, where organizers can create and manage prioritized recognition terms for multilingual meetings.
For workflows using the AI Speech and Captions modality, we recommend using Word Priority to improve recognition accuracy for technical terms when a predefined Glossary translation is not available, or when the term should still be translated naturally across supported languages.
For workflows using the AI Captions Only modality, we recommend using Word Priority to improve recognition accuracy for proper names, brand or product names, and other important terminology spoken during multilingual discussions.
Supported Languages
This feature is supported in all 49 speaker languages.
Arabic
Basque
Bulgarian
Cantonese
Catalan
Chinese
Croatian
Czech
Danish
Dutch
English (US, UK)
Estonian
Filipino
Finnish
French (France, Canada)
Galician
German
Greek
Hebrew
Hindi
Hungarian
Indonesian
Irish
Italian
Japanese
Khmer
Korean
Latvian
Lithuanian
Malay
Maltese
Marathi
Mongolian
Norwegian
Persian (Farsi)
Polish
Portuguese (Portugal, Brazil)
Romanian
Russian
Slovak
Slovenian
Spanish (Spain, LatAm)
Swedish
Tamil
Thai
Turkish
Ukrainian
Vietnamese
Welsh
Troubleshooting
Issue / Task
Possible Cause
Recommended Action
Glossary terms are not translating as expected
Term format does not match the spoken term
Verify the Glossary entry matches the exact term expected to be spoken during the meeting
Do Not Translate terms are still being translated
Alternate spelling or variation not added
Add additional variations, plurals, or acronyms as separate DNT entries
AI translations appear inconsistent across languages
Multiple translation preferences configured for the same term
Use only one preferred translation per Glossary entry
Acronyms are not recognized correctly
Acronyms include punctuation
Add acronyms without periods or special characters (example: “ABC” instead of “A.B.C.”)
AI customization changes are not applied
Meeting already started
AI Customization settings should be configured before the meeting begins
Automatic Language Detection allows KUDO AI to automatically detect the language being spoken during the meeting without requiring participants to manually switch speaking languages.
Performance may vary depending on speaker clarity, microphone quality, overlapping speech, and rapid language switching.
Yes. Translation Languages may be updated after the meeting becomes live. Speaker Languages and AI Customization settings cannot be modified once the meeting begins.