Transcribe Bulgarian audio to text

Bulgarian is written in Cyrillic and is, together with Macedonian, unusual among Slavic languages: it attaches its definite article to the end of the noun and has largely dropped noun cases. For a transcript that means word endings carry more meaning than in Russian or Polish, so a single wrong ending can change a sentence. Reading the editor with the audio running alongside is the quickest way to catch those.

Typical Bulgarian recordings include university lectures, podcasts, municipal meetings and oral-history interviews with older speakers from rural regions. The last group is the most demanding: regional pronunciation and local vocabulary can differ from the broadcast standard, and recordings made on a phone in a kitchen are noisy. Accuracy varies with audio quality, accents and background noise.

Foreign names and brand terms often arrive in Latin letters in the middle of a Cyrillic sentence, and speakers may pronounce them in Bulgarian. Check those in the editor. Each speaker turn carries a timestamp, so you can jump straight to the doubtful passage. Export subtitles as SRT or VTT for video, or a DOCX for review. Originals are deleted within 24 hours and never used for training.