ShotTide中文Sound for your story
Record once, create a private voice, and reuse it for your next narration.
Loading saved audio…
Generate a track to listen, compare and download. Each take keeps its own settings.
Voice cloning creates a reusable speech voice from a recording of an authorized speaker. ShotTide uses MiniMax to create the voice and keeps its selection private to your account. Creating the voice is separate from generating a new script, so you can reuse it across projects.
Upload or record 10-120 seconds of clear speech. For longer uploads, select and preview an excerpt here before applying it. Use one speaker, a steady microphone distance and little room noise. Listen to the sample here before sending it; loud music and overlapping voices make the source less useful.
Give the voice a recognizable name. Optional noise reduction and volume normalization prepare the source for cloning. Once the task succeeds, listen to its preview and select the saved voice in Voice Over.
Adjust speed, pitch, language and emotion when generating narration. A clone preserves a voice direction, not a guarantee of identical performance. Check pronunciation and speaker likeness before sharing.
Choose a recording you are authorized to use. Verify the length and listen for background noise or another speaker.
Confirm the speaker’s permission and review the 50-credit creation cost. The saved task can be reopened after navigation.
Open Voice Over, select the saved voice, and generate a short preview. New narration is charged by text length.
Use 10-120 seconds of clear speech. You can select and preview a local excerpt from an uploaded or saved source up to 10 minutes and 30 MB. Cutting uses no credits; creating a voice is a separate paid action. One speaker and low background noise matter more than an elaborate recording.
Creating a voice costs 50 credits. Later speech uses the normal voiceover rate of 0.2 credits per started 100 spoken characters. Both operations show their cost before submission.
Yes, if your browser supports microphone recording and you grant permission. You can listen before uploading. If recording is unavailable, upload MP3, WAV, M4A, FLAC, AAC, OGG or WebM instead.
The saved voice is selected through your account and checked for ownership on the server. Recordings are sent to the AI provider when you request processing. Only clone your own voice or a speaker who has given permission.
The speech editor offers automatic detection and 24 language hints. Write the new script in the target language; this is not automatic translation. Voice likeness and pronunciation vary, so review a short sample first.
A confirmed failed or canceled task returns its reserved credits. An interrupted response is checked as the same request to avoid charging for an accidental duplicate. Keep the task until its result is known.