Burmese audio to text: what to expect
Burmese has little training data in the speech model, so treat the result as a rough draft. Clear, slow speech from one person gives the most useful text. Expect errors in names, in tones (which change meaning in Burmese) and in fast conversation. Noise, music and overlapping voices make the draft much weaker.
Upload the recording as it is: MP3, M4A, WAV, OGG, OPUS, FLAC, a voice message or a phone recording. You can also record from your microphone in the upload box. Nothing needs converting first.
The transcript opens next to a player that highlights the current line. Click a line to hear it again, correct the word in place, and your changes save automatically.
Myanmar script, Unicode and Zawgyi
The transcript is written in Myanmar script using standard Unicode. That matters because many older phones, websites and documents used the Zawgyi font encoding, which looks the same on screen but stores characters differently. On a device still set up for Zawgyi, our Unicode text can display as broken stacks of characters. Modern Android, iOS, Windows and macOS display Unicode Burmese correctly.
- Burmese does not put spaces between words, so lines can look like long runs of text; that is normal.
- Stacked consonants and medial signs are the characters most likely to come out wrong; check them during review.
- Burmese punctuation such as ။ and ၊ may be missing or replaced by Latin punctuation.
Spoken and formal Burmese, and other languages
Burmese has a spoken register used in conversation and a literary register used in news and formal writing; for example, the sentence ending တယ် in speech corresponds to သည် in writing. The model writes what it hears, so a news broadcast and a family call will look different. English words are common in business and technology talk and may come out in Latin letters or in Myanmar script.
Shan, Karen, Kachin, Mon and other languages of Myanmar are not in the language list, and recordings in them will not transcribe correctly with Burmese selected.
Who transcribes Burmese
Typical files are voice messages from family in Myanmar, interviews with migrant workers and refugees in Thailand, Malaysia and the US, Buddhist dhamma talks, news clips, and research or NGO interviews. Many users treat the draft as preparation for a translator who knows the language.
Export TXT, DOCX, PDF, CSV, SRT or VTT, with timestamps and speaker names switched on or off per export. Speaker labels are available on paid plans or with a Minute Pack.
For video files, see the Burmese video to text page.