Forget about listening to long and boring voice recordings and process them manually.
VoiceCue lets you find sentiments, tags, entities, and actions in your voice recordings like a DJ.
The app workflow is as simple as uploading your voice recording, selecting which type of analysis to perform and clicking on the generated cues in the list to instantly navigate to its exact position in the voice recording.
- Voice recognition - AI based solution by Deepgram
- General stats - an overview about voice recording
- Sentiment analysis - positive and negative word detection
- Word cloud generation - most used word classification
- Entity name recognition - categories such as person, place, etc
- Activity tracking - find actions in past, present or future
- Interactive transcript - see progress or click to control it
- Speaker detection - total number of speakers in recording
- Cue word usage - transcript samples for better context
- Custom search - extended ability to query for cues
- Waveform preview - see the dynamics of voice, identify silences
- Audio controls - play, pause, fast forward and backward
- Drag and drop support - drop audio in the file select area
- Upload MP3 files - most commonly used audio format
- Progress loaders - Improved UX for loading transcripts
- Fully responsive - works fine on mobile and tablets
- Colorful UI - for easier interaction and highlighting
NextJS - React application framework
Deepgram - for AI based speech recognition
compromise, sentiment - for text processing
react-tagcloud - to generate word cloud
react-tabs - for navigation panels
react-drag-drop-files - for drag and drop support
wavesurfer.js - to generate the audio waveform
GitHub - to host the code
Vercel - to deploy the project
ESLint, prettier - for linting and code formatting
VoiceCue is an open source project. The feature requests are welcome.
The project is under the terms of MIT license.