Any2Text changelog
Here we publish meaningful changes to the service: new tools, speech-recognition improvements and interface updates.
Lyrics from audio and MP3
Turn a song into readable lyrics: Any2Text transcribes the vocals with timestamps and exports to TXT, DOCX or SRT — handy for covers, karaoke and subtitles.
Free to start, no signup
Transcribe your first minutes of audio or video for free without creating an account — same AI accuracy (up to 98%), timestamps and export on the free tier.
Subtitle export (SRT & VTT)
Generate ready-to-use captions from any video and download them as SRT or VTT with accurate timestamps — add them to YouTube, Premiere or CapCut in one click.
AI translation in 50+ languages
Translate your transcript into English and other languages before exporting — useful for foreign-language interviews, meetings and subtitles.
Faster processing for long files
Reduced the average transcription time for recordings longer than an hour.
Speaker identification
Any2Text automatically detects and labels different speakers, which makes transcripts of interviews, podcasts and meetings easier to read.
SRT timing on long recordings
Fixed timestamp drift in SRT exports for recordings longer than 60 minutes.