One English setting for every variety
There is no accent menu to pick from. English is one language in the speech model, and it has heard a very large amount of it, from many countries. Clear English speech gives a strong draft whether the speaker is from Texas, Glasgow, Mumbai, Lagos, Sydney or Toronto. What lowers accuracy is usually the recording, not the accent: distance from the microphone, crosstalk, background music and phone-line audio.
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.
What to check for each variety
Each accent has its typical trouble spots:
- American and Canadian English: brand names, acronyms and fast filler-heavy speech in casual calls.
- British and Irish English: regional accents such as Scouse, Geordie or Glaswegian on noisy audio, and local place names. Spelling may mix American and British forms (color and colour), so standardise during review.
- Indian English: personal and place names, numbers said as lakh and crore, and Hindi or other Indian-language phrases inside an English sentence.
- Nigerian and other West African English: names, local expressions, and Nigerian Pidgin, which is not a separate option and is written with standard English spellings where the model can.
- Australian and New Zealand English: slang and shortened words, and Maori place names in New Zealand recordings.
Non-native speakers and mixed languages
Meetings, classes and interviews often include people speaking English as a second or third language. Most accented English is transcribed well; the model mainly struggles when pronunciation is far from typical and the audio is also poor. Keep English selected for these files.
When speakers switch into another language for whole sentences, the English setting may garble those parts or render them as English. If a recording is mostly Hindi, Spanish or another language with some English, pick that language instead. Auto-detect is available, but setting English yourself avoids a wrong guess on files that open with music or a greeting in another language.
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.
Common English recordings and exports
People bring research and job interviews, lectures, podcasts, sales and support calls, voice memos and meetings recorded in Zoom or Teams. International teams use it for calls where accents vary from speaker to speaker.
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 and public video links, see the English video to text page.