๐Ÿ’ก Inspiration Interviews can be stressful, especially when practicing alone. We wanted to build an AI-powered tool that helps users simulate real interview environments using voice, giving them feedback, transcripts, and personalized insights to boost their confidence and communication skills.

โš™๏ธ What it does AI Voice Mock Interview conducts realistic mock interviews using voice input. It listens to the user, responds like a real interviewer using Google Gemini, provides real-time feedback, generates transcripts, and gives improvement suggestionsโ€”all in an engaging interface.

๐Ÿ› ๏ธ How we built it Frontend: React.js, Next.js, Tailwind CSS, Shadcn UI

Backend & Auth: Firebase

Voice & AI: Vapi AI for real-time voice input/output, Google Gemini for AI interview logic

Validation & Types: TypeScript + Zod

Deployment: Vercel

๐Ÿšง Challenges we ran into Integrating Vapi AI with custom voice commands and Gemini responses

Maintaining voice-to-text accuracy in real-time

Managing state between user responses and AI feedback

Designing a modern UI that feels intuitive and non-intimidating

๐Ÿ† Accomplishments that we're proud of Fully functional real-time voice interview experience

Accurate feedback system and meaningful AI questions

Clean and responsive UI for both mobile and desktop

End-to-end working demo deployed live with great user feedback

๐Ÿ“š What we learned Real-time voice interaction and streaming is complex but powerful

Combining Vapi and Gemini opens up powerful conversational AI use cases

Designing UX for mock interviews requires balancing realism and user comfort

Type safety and validation with TypeScript + Zod improved development speed

๐Ÿš€ What's next for AI Voice Mock Interview Support for different job roles and question categories

Multilingual support for diverse users

Resume analysis + behavioral feedback

Dashboard to track user progress over time

More natural voice tones and customizable interviewer personalities

Built With

Share this project:

Updates