Generate new speech from your voice sample
Upload a voice recording, enter the words you want spoken, and generate a WAV file. Listen to the result before you use it.
By JayMiOffice AIUpdated
Sign-in and credits are required for AI processing. Review the displayed cost before starting.
Upload a voice recording, enter the words you want spoken, and generate a WAV file. Listen to the result before you use it.
How It Works
Start with a short recording
Use a recording of your own voice or a voice you have permission to use. Choose a short passage with one speaker so you can hear the voice clearly.
Upload one MP3, WAV, OGG, FLAC, M4A, or AAC file, up to 25 MB. Start with a clear 5–10 second voice sample.
Give the voice something new to say
Enter your script
In Text to Speak, enter up to 4,000 characters. For example, try a short introduction before preparing a longer narration.
Set the pace and process
Use Speed to choose a pace from 0.5× to 2.0×, then click Process. Keep your first script short enough to review the pronunciation and pacing.
Listen before you download
Play the result and check names, numbers, and pauses. Download the generated speech as a WAV file when you are ready.
Treat the first result as a take to review
Your reference recording and written script serve different purposes: the recording supplies the voice sample, while the script supplies the new words. If a line sounds wrong, revise that line and generate another take.
Common questions
Yes. Upload your reference audio and enter the words you want spoken in Text to Speak.
Noise, room echo, music, or multiple speakers can weaken the voice reference. Use a clear recording of one speaker at a steady level and review pronunciation in the generated result. A cleaner sample helps, but does not guarantee an exact reproduction.
Learn More
Continue with a related tool
Choose another step for your project. Each tool shows its supported inputs and processing options before you start.
Create a narrated video: Create a voice track, animate a portrait, and add subtitles.