Week 3 MVP Development #
Project name: VoiceDiary
Code repository: link
VoiceDiary is an AI-powered voice journaling tool that analyzes tone, emotion, and key themes in spoken entries. It generates personalized emotional insights and well-being suggestions based on recorded reflections.
Team #
| Team member | Telegram alias | Innopolis Email | Responcibilities |
|---|---|---|---|
| Dziyana Melnikava | @meldilen24 | dz.melnikava@innopolis.university | PM, Frontend |
| Anastasia Kuchumova | @n_rngk | a.kuchumova@innopolis.university | Frontend, UX/UI |
| Dzhamilia Fatkullina | @jam11a | d.fatkullina@innopolis.university | ML |
| Elina Kuzmichyova | @lin_anile | e.kuzmichyova@innopolis.university | ML |
| Olesia Novoselova | @doiwannaknoww8 | o.novoselova@innopolis.university | Backend |
| Danil Davydyan | @chocop | d.davydyan@innopolis.university | Backend |
Implemented MVP features #
- Emotion recognition based on voice
- Summary for user voices
- User frendly design: minmum effort functionality
- Pause and resume recording without data loss
- Visual feedback while recording: timer and waveform
Demonstration of the working MVP #
Frontend #
Responsible: Dziyana Melnikava, Anastasia Kuchumova
Link to Frontend contribution: CLICK TO SEE
Backend #
Responsible: Olesia Novoselova, Danil Davydyan
Link to Backend contribution: CLICK TO SEE
Swagger demonstartion: CLICK TO SEE
ML #
Responsible: Dzhamilia Fatkullina, Elina Kuzmichyova
Link to ML contribution: CLICK TO SEE
Link to the training code: CLICK TO SEE
Link to model artifacts: CLICK TO SEE
Individual commitments #
| Team member | Telegram alias | Contribution | Link |
|---|---|---|---|
| Dziyana Melnikava (Lead) | @meldilen24 | Connected frontend with backend, Finished adaptive design of Sign-Up and Loging page, Made custom Error handling, updated README | Commits, README |
| Anastasia Kuchumova | @n_rngk | Made results page, Improved UX-UI by adding above microphone clues and a loader, Updated Dockerfile | Commits |
| Dzhamilia Fatkullina | @jam11a | Add new emotion recognition model, added multilingual support and translation, write report | Commits, README |
| Elina Kuzmichyova | @lin_anile | Train Wav2Vec model, add model for text summary, write report | Commits, Models, README |
| Olesia Novoselova | @doiwannaknoww8 | Implemented the logic for voice data transfer, improved integration between the Go API and the Python ML service, set up Docker Compose | Commits, Swagger Demo |
| Danil Davydyan | @chocop | Improved communication between the backend and ML, established communication between the Go API and Python API | Commits |
Plan for Next Week #
ML #
- Research emotion recognition model based on text
- Research LLM model to add support comments based on user voices
Frontend #
- Finish Authorization page and connect it with backend
- Create Hompage
- Limit record length for unauthorized users
Backend #
- Service logs
- Secure storage of user data
- Performance measurements under load
- Authorization
Confirmation of the code’s operability #
We confirm that the code in the main branch:
- [✓] In working condition.
- [✓] Run via docker-compose (or another alternative described in the README.md).