CloneVoice

以下英文版少于 **4,000 个字符**，可直接粘贴到 App Store Connect： --- CloneVoice is an offline voice cloning and text-to-speech app. Record your own voice, or import a recording of someone whose voice you have permission to use. Enter text to generate speech with the reference voice's characteristics. Clone a voice from a short recording Record or import 3–10 seconds of clear, natural speech to establish a reference voice. No specific script is required. For better results, record in a quiet environment with minimal background music and noise. Turn Chinese or English text into speech Enter short passages in Chinese or English and generate speech using the reference voice. Preview written copy, read short passages aloud, or create voiceover material for short videos. Run offline, process on your device The voice model is included with the app. Voice cloning and speech synthesis run entirely on your device, without uploading recordings or drafts to the cloud. No account is required to start creating. Preview your creations and export WAV audio Play back generated speech directly in the app. Export WAV files through the system share sheet to save, share, or continue working in other audio editing tools. Manage your drafts and creations Drafts are saved automatically, and generated audio is kept for later playback. Continue editing your text, delete the saved voice, remove individual creations, or clear all local data managed by the app. Usage notes • The app is designed for short passages, with up to 30 seconds of synthesized speech per creation. If your text is too long, shorten it and try again. • Voice similarity, audio quality, and generation speed depend on the reference recording, input text, and device performance. • The model requires substantial memory. A newer device with at least 8 GB of RAM is recommended. • The interface supports Simplified Chinese and English and follows your device's language settings at launch. Use only your own voice or a voice you have explicit permission to use. Do not use the app to impersonate others, commit fraud, or bypass voice authentication. Exported audio includes an audible AI marker and AI metadata. Keep these labels when sharing and let listeners know the content is AI-generated. Support: hezhuangjs@gmail.com