AI Speech to Text creates a transcription from an audio file. Review the result against the recording before treating it as an accurate record.
How do I transcribe a file?
- Open AI Speech to Text and select Generate AI transcription.
- Give the transcription a recognizable name.
- Choose a file using the upload field. Follow the permitted formats and maximum file size shown on your screen; the allowance can vary by plan.
- Review any Advanced settings you need, then select Create.
- Open the resulting transcription and correct important names, numbers, punctuation, and unclear passages before reusing the text.
Which files are accepted?
The current upload form lists MP3, MP4, MPEG, MPGA, M4A, WAV, and WebM. Use the size limit actually shown in your account. A larger plan allowance does not make an unsupported file type valid.
How can I improve the result?
Use clear speech with limited background noise and avoid overlapping speakers when possible. If a file is too large, prepare a shorter or appropriately compressed copy and check it is still understandable. Only upload recordings you are authorized to use.
Why is the transcript incomplete or wrong?
Poor audio, unusual names, multiple speakers, and missing context can affect transcription. The output is a draft, so verify important passages directly. See AI troubleshooting if creation fails or the allowance appears incorrect.