A simple recording journey
Built the interface logic for microphone recording, audio playback, processing states and text download. The current recognition step returns demonstration text.
Audio / AI prototyping
From spoken words to useful text.
Two complementary prototypes exploring audio capture, local speech recognition and text export: a React recording interface and a Python transcription service.
Inside the projectProject stage
Component prototypes
Contribution
Recording-interface prototyping and local speech-processing implementation.
Disciplines
AI & automation · Websites & digital products
01 / The challenge
Turning a conversation into reusable notes involves several steps: capturing the audio, preparing its format, recognising speech and returning a readable file. The work explored those steps as separate components.
02 / The work
Built the interface logic for microphone recording, audio playback, processing states and text download. The current recognition step returns demonstration text.
Implemented a separate Flask service using Whisper for Russian-language audio. It accepts uploaded audio, prepares the format and returns text with a downloadable transcript.
Separated the user-interface experiment from the speech-processing service. Connecting the components and aligning their audio formats are the next implementation steps.
03 / What the work produced
A recording-interface prototype and a separate local transcription implementation, providing the components for further integration and evaluation.
The recording interface and local transcription service are separate prototypes. Their integration remains the next step; the illustration shows the intended flow.

Start a conversation
A brand, a website or a campaign. Tell us what you have in mind.