Workroom's AI workspace for video, image, and voice generation includes Voice Studio — a text-to-speech tool that runs on one of two models. Workroom (QwenTTS) is the built-in engine that covers everyday voiceover needs with a broad voice library. ElevenLabs is a third-party integration with a different voice catalog for specific production workflows.
Workroom
Workroom is Voice Studio's native text-to-speech model. It produces clear, natural-sounding speech from any text input. Common uses: product videos, social media content, app narration, and podcast intros.
The model includes a library of pre-made voices spanning a wide range of ages, genders, tones, and use cases. Use the Use Case, Tone, and Age filters at the top of the voice library to narrow down candidates. Each voice has a short audio sample — click the play button to preview before selecting.
ElevenLabs
ElevenLabs is a professional TTS integration available as an alternative when your project requires a voice profile not found in the Workroom library. It covers use cases like narration, character work, and voiceover production.
The interface is identical to Workroom — the same Pre-made / My Voices / Create Voice tabs, the same four filters. Voices created under ElevenLabs are stored separately from Workroom custom voices.
Switch models
Open the Model dropdown in the top-left corner of Voice Studio and select the model you want. The voice library updates immediately to show voices from that model. Your text input stays in place.