Text to Speech with Target Voice Selection
Type or import a script, choose a target voice, preview the voice, adjust speaking speed, select an output folder, and generate the spoken result.
Create spoken audio from text, transform an existing recording into a selected voice, or build a reusable custom voice from a short authorized reference sample.

GiliSoft VoiceLab is built around three connected jobs. Use Text to Speech when a script needs a natural voice, Voice Conversion when an existing recording should be spoken by another selected voice, and Custom Voice when you need to create and reuse an authorized voice identity.
You can search the voice library, filter voices by language or category, preview samples, save favorites, and return to your own custom voices. The task list keeps generated files, point use, status, details, and retry actions visible after processing.
VoiceLab is a focused AI voice product. It is different from the basic Text to Speech tool inside Audio Toolkit because it also covers voice conversion, custom voice cloning, a broader target-voice library, and reusable voice management.

The page follows the actual controls shown in VoiceLab, from preparing source material to choosing a voice, generating audio, and reviewing completed tasks.
Type or import a script, choose a target voice, preview the voice, adjust speaking speed, select an output folder, and generate the spoken result.
Upload source audio or video, or record directly from a microphone, then choose a target voice and generate a converted voice result from the same screen.
Create a voice clone from an authorized audio or video reference or a short recording. Add a name, language, and tags so the voice is easier to find later.
Browse a broad voice collection, filter by language and category, preview individual voices, add favorites, and send the selected voice directly into a generation task.
Keep recent voices, favorites, and custom voices organized. The task view records file name, task type, target voice, point use, status, details, and retry options.
The interface shows the current balance and estimated point rule before a text-to-speech or voice-conversion task is generated, helping users plan usage before processing.
Choose a task below to inspect the full interface. Click the large screenshot to open a larger preview.

Enter or import text, choose a voice, adjust speed, choose an output folder, and review the displayed point rule before generating audio.
Generate narration drafts, convert an existing recording to a selected voice, and keep reusable voices organized for lessons, demos, explainers, and training material.
Search voices by language, preview options, save favorites, and prepare alternate spoken versions without rebuilding the project around a separate voice tool.
Use a written script, uploaded recording, video source, or direct microphone recording as the starting point for a generated voice task.
Create and label a reusable voice from approved reference material, then return to My Voices when the same identity is needed for another project.
Import the text, choose and preview a suitable voice, adjust speed, select the destination, and generate the spoken result.
Add an audio or video source or record directly, choose the target voice, and keep the generated task visible in the task list.
Create a voice from an authorized reference sample, give it a clear name and tags, and retrieve it later from My Voices.
It combines text-to-speech generation, audio and video voice conversion, custom voice cloning, voice-library browsing, and generation task management in one Windows application.
Yes. Create Voice Clone accepts an authorized reference audio or video file or a short recording, then lets you name and organize the resulting custom voice.
Yes. Voice Conversion accepts uploaded audio or video and direct microphone recording, then generates the result with the selected target voice.
The voice library shown in the product supports language and category filters, search, preview, favorites, and direct voice selection.
The Tasks screen records the file name, time, task type, selected voice, point use, status, and detail or retry actions.
Use a clear recording you own or are authorized to use. The built-in recording screen shown in VoiceLab supports a short reference capture and provides fields for the voice name, language, and tags.
Start with text, an existing audio or video recording, or an authorized reference voice, then use GiliSoft Points when you are ready to generate the result.