Audio is the medium I grew up in — I have aphantasia, so sound became my primary creative space early. I learned by practice and from the best local producers I could find, years before the internet made that easy. 20 years of producing and mixing music, 2+ years engineering audiobooks, and now building the tools that automate the studio work around it.

By day I record, direct, and edit audiobooks at DigiRead — planning sessions, recording narrators, helping with pronunciations and character voices, cutting and aligning files. The rest of the time I build the pipelines that scale the output: an Estonian-language audiobook TTS platform on the ElevenLabs API, NLP processing, QA tooling, and reporting automation. Most of these builds leverage AI and are built for my own production use — functional and effective, not polished open-source code, but they work at scale every day.

Before audio, I ran a night club (Ulme, 2012–2015) and an open-concept kitchen (Puljongibaar, 2017–2020). During COVID I went deep into audiobooks and was struck by how much knowledge people could absorb through audio. Got the studio role through acquaintances from my earlier music production years, and now the coding and AI skills from my degree at Kood/Jõhvi merge naturally with the audio work.

• DigiRead Studio — full-stack Estonian audiobook TTS platform (manuscript → ready-to-deliver audio) built on the ElevenLabs API.
• Studio tooling — narrator alignment & pause detection, run-tracking/reporting, and a real-time reading-position tracker.

• Languages: Python, JavaScript/TypeScript, Go, SQL
• Web: Next.js, React, FastAPI, Node.js
• Data/Infra: Supabase (PostgreSQL, Auth, Storage, RLS), Turborepo
• AI/Audio: ElevenLabs TTS, Whisper/WhisperX, NER, text normalization, ffmpeg/ffprobe
• DAWs: Ableton Live, Adobe Audition, Logic Pro

• Estonian (native), English & Russian (fluent).
• Bandcamp · YouTube · Dragan Volta