About the service
I take video and bring it to finished subtitles ready to upload immediately. First, I run the audio track through speech recognition — it provides rough text with approximate time splits. Then manual work begins: I compare each line with the video, fix timing where the engine drifted from the speech, and rewrite phrases the recognition got wrong — names, terms, accents, overlapping voices, background noise near the microphone. I don't deliver raw machine text as finished subtitles. Before handing over the file, I watch the entire video with captions on and ensure lines read comfortably, don't race ahead or lag behind the spoken words, and line breaks don't split phrases mid-sentence. I work with interviews, educational videos, webinars, video podcasts, corporate and advertising videos — project length and topic don't matter, only the source audio. What's included: — speech recognition and rough timing sync — manual correction of disputed and unclear sections — export in formats for video platform upload — control of reading speed and line length on screen — one round of revisions after delivery
