AI models on Craze
Every video, image, voice, music and sound model Craze can direct, with what each one makes and what it costs.
Every video, image, voice, music and sound model Craze can direct, with what each one makes and what it costs.
Starts at$0.01 per voice
Kuaishou's voice creation route. Upload one voice sample to make a reusable profile; this operation does not generate video or narration.
Upload one authorized voice sample for a character, client, or yourself.
Voice creation does not synchronize audio and video.
Create one voice profile per character in separate calls.
This route creates a profile from audio.
Craze creatively jams, researches and plans with you.
Craze generates anything imaginable. Video, audio, music, motion graphics and more.
Craze watches your footage and builds a final cut. With a real editor if you need it.
Every model here runs on every Craze plan, including Free, and Craze can use a different one for each shot.
| Kling Voice CloneKuaishou | ElevenLabs v4ElevenLabs | Gemini 3.8 Flash TTSGoogle DeepMind | Gemini 3.8 Flash Lite TTSGoogle DeepMind | |
|---|---|---|---|---|
| Makes | Audio to audio | Text to audio | Text to audio | Text to audio |
| From | $0.01 per voice | $0.085 per 1,000 characters | $0.032 per 1,000 characters | $0.022 per 1,000 characters |
| Task | Inputs | Resolution | Aspect ratios | Sound |
|---|---|---|---|---|
| Voice cloning | Voice sample | Follows the source | Follows the input | None |
$0/ month
$19/ month
$59/ month
$199/ month
Kling Voice Clone creates a voice profile from one audio sample. On Craze, voice creation and video generation are separate operations; this page covers the profile creation cost.
Save a character voice once, then choose a compatible generation tool for the video or speech that uses it. Cloning alone does not render a clip.
Upload one authorized voice sample for a character, client, or yourself. The resulting profile can be reused in compatible workflows.
Voice creation does not synchronize audio and video. Use a separate video or lip-sync operation when your project requires timed dialogue.
Create one voice profile per character in separate calls. This route accepts one sample per call and does not render a multi-character scene.
This route creates a profile from audio. Whether it can be used for a particular performance depends on the separate generation route you choose.
You pay $0.01 per voice for a voice profile. Each additional character voice requires another creation call. Video and speech generation are priced separately.
Animated characters need consistent voices across episodes. Clone each character's voice once and use it for every scene they appear in.
Cutscenes with multiple characters may need separate voice profiles and a separate video generation or lip-sync workflow.
Create a signature voice for your brand character or spokesperson. Clone it and use the same voice across all videos to build brand recognition.
Create a reusable voice profile for compatible localization workflows. This route itself does not translate or narrate.
Seed Audio 1.0 generates speech from text with optional audio or image references. Kling Voice Clone creates a voice profile from one audio sample. For synchronized video, use a separate video or lip-sync model.
For text-to-speech, Seed Speech and MiniMax Speech 2.8 HD accept text and a voice ID. Kling Voice Clone creates a profile; it does not combine voice cloning with video generation in one call.
Kling Voice Clone does not generate audio and video together. Create a profile here, then use a compatible video and lip-sync route if you need the character's mouth to match dialogue.
For character-driven content, plan the voice and visual synchronization as separate steps. This profile-creation operation alone does not determine timing.
On Craze, you can create separate profiles from different samples. Scene variations belong to a separate video generation task.
For multiple speakers, create each character's voice profile in a separate call. This route accepts only one voice sample per call.
Character placement and screen time are video-generation concerns, not outputs of this voice-creation route.
Use a clear, single-speaker audio sample. The Craze catalog does not specify a 10-to-30-second limit or expose a clone-quality preview step for this route.
Describe character appearance and movement when you use a separate video-generation model. Voice creation does not take a video scene prompt.
Refining the visual scene requires a separate video generation; it does not change the saved voice profile.
This route creates a profile; it does not generate a singing performance or a music video. Choose a compatible generation model for those tasks.
For episodic content (animated series, game dialogue sequences, branded character series), clone the character voice once and use it for every episode. The voice stays consistent because the clone profile is stable. Viewers recognize the character across all content. This is more efficient than re-cloning or re-recording for each episode.
Reuse a voice profile in compatible generations where it is available. Each additional profile is a separate voice-creation operation.
This voice-creation route has no language-conversion or translation option. For localization, use a separate dubbing or text-to-speech model.
Hand Craze your audio and say what you want, and it runs Kling Voice Clone for you. Every Craze plan, including Free, runs Kling Voice Clone and every other media model, and Craze can use a different one for each job. Each voice is saved as a reusable profile, ready for the voiceovers you make next. Generations come out of your plan's monthly credits, from $0.01 per voice, with top-ups when you need more.
Character-driven animated shorts and series: Animated characters need consistent voices across episodes. Game cinematics and dialogue scenes: Cutscenes with multiple characters may need separate voice profiles and a separate video generation or lip-sync workflow. Brand mascots and spokesperson videos: Create a signature voice for your brand character or spokesperson. Multilingual video localization: Create a reusable voice profile for compatible localization workflows.
Kling Voice Clone creates a reusable voice profile from one voice sample. On Craze this operation does not produce speech or video.
Kuaishou released Kling Voice Clone as part of the Kling 2.6 video model on December 3, 2025, enabling simultaneous audio-visual generation.
You pay $0.01 per voice. The cost calculator on this page shows the total before you generate.
every Craze plan, including Free.
A generated voice profile can be selected in a separate compatible generation. This page prices only the voice creation step.
Upload one clear voice sample. Craze does not accept multiple speakers or separate character samples in a single voice-creation call.
Upload a single voice sample and ask for Kling Voice Clone. The result is a voice profile, not a video or a narrated script.
