Dubbing for Floating Cloud Dreams

What is the voice casting for Float Cloud Dream?

Float Cloud Dream Dubbing is a web-based AI voice generation tool that can convert text into natural-sounding speech.

The platform also offers multi-person conversations, batch voiceovers, voice cloning, speech conversion, subtitles, AI-generated music, and AI art.

Main functions

• Combine the input text into a downloadable audio file.
• More than 400 neural network timbres are available.
• It covers more than 140 languages, regional variants, and Chinese dialects.
• Adjust the speech pace, tone, volume, and emotional style.
• Set separate voices for multiple characters and generate complete dialogues.
• Process long texts in batches via asynchronous tasks.
• Create cloned timbres using short vocal samples.
• Convert existing recordings to other timbres.
• It identifies audio and video content and generates subtitles.
• Calibrate and convert SRT, ASS, and VTT subtitles.
• Generate music or images using prompts.

Text to speech

After entering text and selecting the language and voice tone, the user can generate speech without the need to install any client software.

HD timbres can be adjusted to reflect the emotion conveyed in the text, but automatic emotion detection may still differ from the intended intent behind the creation.

Language and Chinese dialects

Officials currently state that it supports over 140 languages and dialect variations, as well as offering more than 400 AI voice tones.

• Chinese includes Mandarin, Cantonese, Sichuanese, Shanghainese, and others.
• Some pages also list the dialects of Northeast China, Henan, Shaanxi, and Shandong.
• English is spoken in regions such as the United States, the United Kingdom, Australia, and Canada.
• It also includes Japanese, Korean, French, German, and Spanish.
• The timbres and emotional styles available in different languages are not exactly the same.

The names of dialects represent available models; they do not ensure that every nuance related to accent, age, or region can be accurately reproduced.

Limit on single generation and length of text

For long texts, it is recommended to divide them into chapters; first use short paragraphs to confirm the pronunciation, characters involved, and speaking pace, and then create the complete task.

Voice acting for multi-person conversations

In multi-person conversations, it is possible to assign separate vocal tones, speech speeds, pitches, and styles to different characters, with the system automatically synthesizing a complete conversation.

1. Organize the script in the format “character name: English colon: dialogue”.
2. Ensure that the same character always uses the same name.
3. Save separate sound settings for the narration and for each character.
4. Generate a few dialogues to check the role switching.
5. Handle the pronunciation of homophones, numbers, and special names.
6. Once verified, the complete radio drama or audio dialogue is generated.

This feature is suitable for audiobooks, radio dramas, interview presentations, classroom dialogues, and short video stories.

Batch voiceover

The batch creation feature is designed for audiobooks, series of courses, and large amounts of advertising copy, and it can be executed asynchronously in the background.

• Each task can handle up to 100,000 characters.
• It is possible to import bulk content in the form of text or tables.
• It allows multiple tasks to be submitted, with the option to check their status later.
• Quality options such as standard, high definition, and ultra-high definition are available.
• After completion, it should be listened to and downloaded as soon as possible.

The asynchronous generation time is influenced by the length of the text, the task queue, and the load on the model; therefore, it is not possible to guarantee a delivery time in fixed seconds.

Voice cloning

Voice cloning learns the acoustic characteristics from 5 to 30 seconds of human voice samples, and then uses that voice timbre to read new text.

1. Only use one’s own voice or samples for which explicit permission has been obtained.
2. Choose a quiet environment to record clear vocals without echoes.
3. Avoid including music, multiple voices, or sensitive information in the samples.
4. First, use risk-free short sentences to test similarity and stability.
5. It should be stated at the time of publication that the content was generated by AI, in order to avoid misleading the audience.
6. After use, delete the unnecessary samples and generated files.

Cloning someone’s voice without consent can violate their personality rights, privacy, or rights related to performance, and it may also be used for fraud.

The difference between voice conversion and voice cloning

Voice conversion preserves the rhythm and expression of the original speech, while ordinary TTS generates speech from text anew.

Subtitle generation and calibration

The platform can identify speech from audio or video, export SRT subtitles or plain text, and supports multilingual translation.

• Adjust the overall timeline offset of the subtitles.
• Merge subtitles that are too short or consecutive.
• Split long subtitles that are difficult to read.
• Convert formats between SRT, ASS, and VTT.
• Check names, places, numbers, and technical terms.
• Import it into CapCut or another professional video editing software to continue editing.

Automatic subtitles may miss words or have timing issues; they need to be checked sentence by sentence before being released officially.

Fine control of SSML

Users familiar with voice-over parameters can use SSML tags to control pauses, stress, speech speed, and tone.

AI music and AI art generation

In addition to voice, the website also offers independent creative tools such as text-to-music and image generation.

• AI music can generate audio based on style descriptions and lyrics.
• Music can be used for video soundtracks, podcast intros, or to create an atmosphere in games.
• AI drawing supports Chinese descriptions and various visual styles.
• Before editing with reference images, it is necessary to confirm the rights to use those materials.
• The generated results may be similar to existing works or characters.

The commercial use of music and images also depends on the input materials, character images, trademarks, and third-party rights.

Tutorial for dubbing in Float Cloud Dreams

1. Visit the official website; do not download any third-party clients with the same name.
2. Enter content in the text box, or upload TXT or DOCX files.
3. Select the language, timbre, speech speed, pitch, and volume.
4. Names, characters with multiple pronunciations, numbers, and English words are checked first.
5. Use a trial audio text to check whether the timbre and mood are appropriate.
6. Insert pauses, styles, or background music as needed.
7. Click to generate the audio and listen to the full result.
8. For long texts, use the batch generation feature and submit them in sections.
9. For multi-player scripts, first configure the characters and then verify to which segment each character belongs.
10. Download MP3s, subtitles, or other files you need.
11. Before publishing, verify the authorization, facts, pronunciation, and AI identifier.
12. Save the finished products in a timely manner, without relying on temporary server files.

Free use and pricing

The official website currently lists its core features as free of charge, requiring no login, and without any daily or weekly character limits.

The official website offers both sign-in points and membership rewards; these benefits or pricing structures may change in the future, so it is advisable to check the page instructions before using them.

Is registration required?

The fact that it’s available for free doesn’t mean that all account features require no registration; cloud synchronization and various benefits associated with events do require an account.

Commercial licensing boundaries

The official website states that the generated audio comes without watermarks and can be used for commercial purposes free of charge, but users still need to ensure that the way in which it is inputted and utilized is legal.

• Original content can be used in one’s own projects after its validity has been verified.
• For other people’s articles, novels, dialogue, and lyrics, appropriate permissions must be obtained.
• Cloning a sound requires explicit permission from the rights holder of that sound.
• Background music, reference images, and uploaded videos also have their own separate rights.
• Do not use synthetic voices to pretend to be oneself in order to provide endorsements or issue statements.
• Advertising, financial, and medical content are subject to stricter rules.

“The fact that the platform allows commercial use” cannot replace the user’s own verification of third-party copyrights, personal rights, and compliance with industry regulations.

File saving and privacy

Different sections on the official website specify that audio files are retained for either 60 minutes or 10 minutes, resulting in inconsistencies in the official statements.

The safest approach is to download the file immediately after it has been generated, treating the cloud-based file as a temporary one.

1. Do not enter passwords, identification documents, or customer secrets in the text to be processed.
2. The cloned sample records only the necessary content and removes the background dialogue.
3. Obtain permission from participants and copyright holders before uploading meetings and courses.
4. Once generation is complete, download it immediately and check that the file can be played.
5. Clear local history and role configurations on shared devices.
6. Check cloud synchronization and the logout option when no account is in use.

Clients, APIs, and open-source status

The official website warns that apps and computer software with the same name may be unofficial clones, and it is necessary to verify their origin before installing them.

Use sound cloning safely

• Rejecting the cloning of celebrities, colleagues, or family and friends’ voices for impersonation.
• No voice messages intended to prompt transfers, verification codes, or identity confirmation are generated.
• Include clear instructions regarding AI synthesis when releasing it to the public.
• Team projects retain authorization scopes and withdrawal mechanisms.
• If you discover that sounds are being misused, save the evidence and contact the platform.
• For important identity verification, callback and multi-factor authentication are now used.

Usage restrictions

• Free rules, activities, and available models may change due to operational adjustments.
• A single TTS output of over 5,000 characters may be truncated.
• Batch tasks need to be queued, and no fixed completion time is guaranteed.
• Different timbres support varying languages and emotional styles.
• Automatic subtitles, characters with multiple pronunciations, and dialects may be misidentified.
• The official website provides inconsistent information regarding the duration for which temporary audio files are stored.
• The platform does not provide external APIs, so it is not possible to retrieve data through its interfaces on one’s own.

Frequently Asked Questions

What are the main functions of the floating cloud dream voice-over feature?

It offers text-to-speech, multi-person conversations, batch voice recording, voice cloning, voice conversion, subtitles, AI music, and AI drawing.

Is the voice acting for Float Cloud Dream really free?

The official website currently lists its core functions as free and accessible without logging in, but the site also features points systems and member-exclusive offers; the rules may change in the future.

How many characters can be converted per transcription for the Floating Clouds Dream audio?

For standard text-to-speech conversion, a maximum of 5,000 characters can be processed in a single operation; for longer texts, batch tasks can be used, with each task allowing up to 100,000 characters to be processed.

Can the audio generated by Float Cloud Dream’s voice synthesis be used for commercial purposes?

The official website states that it can be used for commercial purposes free of charge, but users still need to ensure that the text, audio samples, music, images, and character designs have proper licensing.

Is there an official app for the dubbing of Float Cloud Dreams?

No, the official website clearly states that there are no apps or desktop software; any client with a similar name should be carefully checked to determine whether it is a fake product.

Does the Floating Cloud Dream dubbing service offer an API?

No, the official website clearly states that no external APIs are available, and calls to any interfaces other than those provided on the site are prohibited.

How long is the generated audio stored on the server?

The official website provides figures of 10 minutes and 60 minutes respectively, and these figures are inconsistent; it is recommended to download the file immediately after it is generated and to keep a backup of it.