Text-to-speech guide
Choosing safe settings
Choose a voice matching the language and review abbreviations, numbers and punctuation. Short sentences and clear punctuation usually create more natural pauses.
Speed affects intelligibility. Generate a short sample first, then adjust wording or spacing when a name or technical term is mispronounced.
Quality checklist
Speech generally needs less data than music. Raising bitrate above the source quality does not restore missing detail. Increase volume carefully because clipping creates harsh distortion.
Keep an untouched copy, trim before converting, and listen to quiet and loud sections after export. Verify duration and use descriptive filenames for different versions.
Privacy and safe handling
Media can reveal faces, voices, locations, documents and private conversations. Review a file before sharing it and remove unnecessary metadata when appropriate. Avoid uploading confidential work to any online processor unless you understand its privacy practices and have authority to do so.
A practical selection checklist
For spoken audio, moderate AAC or Opus bitrates are usually enough. Music benefits from a higher rate and stereo channels. Keeping the original sample rate avoids unnecessary resampling. For archival or editing work, a lossless audio format is useful, but it creates much larger files.