Interview Related Issues
Q: Which platforms can be used?
A: We support all major platforms: DingTalk, Tencent Meeting, Feishu, Teams, Zoom, WeChat Voice, mobile phone calls, web interviews, and more. If you're unsure whether your interview platform is supported, you can contact our customer service for assistance
Q: Why is there no response from the page after a while?
A: If this happens, please contact us
Q: What are the differences between Standard Mode, Light Mode, Coding Mode, System Design Mode, and Detailed Mode?
A:
Standard Mode: Answers are provided using default settings
Light Mode: The AI first summarizes the answer and then provides brief, point-based responses, suitable for quick answers during an interview
Coding Mode: Specifically for coding questions asked by the interviewer during the interview process
System Design Mode: Focused on backend technical interviews where the interviewer asks system design questions, such as "Please design a short URL service system"
Detailed Mode: The AI provides very detailed answers, covering all aspects of the question.
Q: Will wearing glasses cause reflections?
A: It has been tested and there is no reflection!
Q: Can one phone be used to complete the interview assistance? How about one computer?
A:
One phone cannot complete the interview assistance because the phone's microphone can only be used by one application at a time. If the interview software is using the microphone, InterviewDog will not be able to hear it
Using a tablet is also not possible, for the same reason
Using one computer is possible. You can open InterviewDog in one browser tab and use another browser tab or interview software for the interview
Q: What is the principle behind automatic question recognition?
A: InterviewDog continuously listens to the conversation between the interviewee and the interviewer. Once it detects an interview question in the conversation, it will automatically extract it
Q: How to simulate an interview?
A: After filling in the source information, each account will receive free usage time, which can be used for testing. You can act as the interviewer and ask questions directly to InterviewDog, or you can have a voice call with a roommate or friend to simulate the interview
Q: How to choose speaker mode?
A: If you select "Interviewer", InterviewDog will treat all speakers (including yourself) as potential interviewers. It will analyze the speech for possible questions, and if detected, it will extract them automatically. If you select "Candidate", InterviewDog will still record what is said, but it will not automatically extract questions from the conversation.
Q: How to select the audio input method?
A:
If the interviewer's voice is coming from the desktop client, then select "System Audio Mode"
If the interviewer's voice is coming from another webpage, then select "Web Audio Mode"
Q: What kind of conversations will be recognized as questions?
A: Any conversation that could potentially be an interview question will be extracted as a question. For example: "Can you tell me about xxx?", "How was xxx solved?", "Tell me about yourself", etc
Q: The question extraction is too fast, and a new question appears before the previous one is answered
A: You can use the shortcut key "F" or manually click on the "Fix Current Answer" in the bottom left corner of the screen to prevent new questions from overlapping with the current one. Remember to close it when you're done, or it will remain on the current question page
Q: Why is what I say also being recognized as a question?
A: The reason we designed it this way is to allow you to provide additional context if the interviewer's speech recognition is inaccurate. For example, you can repeat the question or confirm it multiple times
If you don't want your speech to interfere, you can switch the answer role to "Candidate." After switching, your voice will still be recognized, but it will no longer be extracted as a question
Q: I want the AI's answers to follow my requirements, such as "Don't make the answer too formal" or "Answer all in a specific format." What should I do?
A: You can use our "Custom Prompt" feature
Q: What should I do if the speech recognition is inaccurate?
A: Our AI model has been trained, so even if the speech recognition is inaccurate, the answers can still be provided accurately. However, the interview scenario is more complex, as factors such as the interviewer's pronunciation, the interview environment, and device quality can affect the accuracy of speech recognition. For solutions, we recommend the following two options:
(Recommendation) Report incorrect recognition: Report cases where speech recognition was incorrect in the interview record. After our review, similar speech will not make the same mistake in the future. Additionally, you will receive a reward for helping correct our speech recognition model
Repeat the pronunciation yourself: During the interview, repeat the pronunciation once more according to the standard pronunciation
Q: What should I do if I need to answer questions during the interview?
A: You can activate our "Writing Assist" feature, which supports remote screenshot capture through the client during the interview and provides real-time answers to written test questions
Last updated
