Transcribe Malay audio to text, free

What makes Malay hard to transcribe

Malay and Indonesian share a common root and much vocabulary, which means an auto-detected transcript can quietly come out in the wrong standard — same sounds, different spelling conventions and different everyday words. Malaysian speech also mixes English far more than Indonesian does.

What your Malay transcript looks like

Malaysian spelling conventions rather than Indonesian ones, with English terms left in English where the speaker used them.

[00:00 → 00:04] Selamat pagi semua, kita mulakan mesyuarat hari ini.

[00:04 → 00:08] Pertama sekali, kita bincangkan keputusan bulan lepas.

An illustration of how Malay transcripts are formatted — script, punctuation and timestamps. Not the output of any particular recording.

How accurate to expect

Standard Malay transcribes reliably. The frequent English mixing in Malaysian speech is handled, but auto-detect may classify the audio as Indonesian.

What usually needs fixing

1. 1 Indonesian spelling conventions applied to Malaysian speech when the language is auto-detected.
2. 2 Arabic-derived religious vocabulary rendered inconsistently between Malay and Arabic spelling.
3. 3 Regional Malaysian accents from the east coast states, which are less represented than Kuala Lumpur speech.

Getting better results

• → Select Malay rather than Indonesian explicitly, or you may get Indonesian spelling for Malaysian speech.
• → Manglish — Malaysian English with Malay particles like lah — transcribes as spoken.

Who transcribes Malay