Overview
Speech to Speech in the Pro Editor lets you record or upload a human voice performance and convert it into any AI voice in the AudioStack library. The output preserves the pacing, rhythm, and emotion of the original recording while applying the selected AI voice.
This is particularly useful when a client or voiceover artist records a reference take, or when you want to rapidly prototype how a script sounds with different voices without having to re-record.
Opening the Generate Speech Panel
In the Pro Editor, click Generate Speech in the top-left of the toolbar
The left panel opens showing your existing speech clips, or an empty state ("No Speeches Yet") if this is a new project
Click + Create New to open the Generate Speech modal
Using the Speech to Speech Tab
The Generate Speech modal has two tabs: Text to Speech and Speech to Speech. Select Speech to Speech.
Step 1 — Add your source audio
You have three ways to provide the source voice recording:
Record — record directly in the browser. Click the red microphone button to start and stop. A live timer shows the duration. Use the settings icon (⚙) to adjust input device if needed.
Library — pick from recordings you've already captured in AudioStack
Upload — upload an audio file from your computer
Step 2 — Select the Output Voice
Under Output Voice, you'll see the currently selected AI voice with its name, language, accent, and personality traits (e.g. Upbeat, Confident, Fast, Informative).
Click Change Voice to browse the full voice library and select a different voice
Use the ▶ button next to the voice name to preview it before committing
Step 3 — Voice Options
Expand Voice Options to access:
Expressive Voice — toggle this on to allow the AI voice to apply more dynamic expression based on the emotional content of the source audio. Leave it off for a more neutral, consistent delivery.
Step 4 — Preview and Generate
Click Preview to hear a short sample of the conversion before committing
When satisfied, click Generate (blue button)
The converted speech clip will appear on the Voices track in the timeline
💡 Tip: If you want the AI voice to mirror the exact pacing of the original recording, keep Expressive Voice toggled off. Turn it on when the source recording has strong emotional cues you want to carry through to the AI output.
Script Text
The Script Text field on the left side of the modal is required for Speech to Speech in Pro Editor. Adding the script text here will help you to find and reuse this speech clip later.
Common Use Cases
Rapid voice comparisons: use one recording and generate multiple versions with different AI voices to present options to a client
Preserving a presenter's rhythm or style: a radio presenter records their style of delivery; Speech to Speech retains that rhythm in the AI voice output
Changing age or gender