Frequently Asked Questions

Yes, all tools are permanently free with no hidden fees and no credit card required. We promise never to charge you a single cent, and we will never charge for basic features in the future.

Because AI voice synthesis and cloning require massive GPU computing power, and server resources are limited. To ensure fair usage for everyone, when too many users are active simultaneously, the system automatically queues new requests and processes them in order. It's like going to the bank — when it's busy, you have to wait in line.

It depends on the current queue length and task complexity. Voice cloning (like IndexTTS 2.0) typically requires longer processing time, with queues possibly taking 1-10 minutes; regular voice synthesis (like Qwen3-TTS) is relatively faster, with queues about 30 seconds to 3 minutes. The page will display your queue position and estimated remaining time in real-time.

You can close the page — queuing won't be interrupted. Once processing is complete, the result will be automatically generated, and you can re-enter the page to download. Of course, you can also keep the page open and wait — the progress bar will update dynamically.

Currently, no paid acceleration services are supported. We insist on being completely free, so everyone is treated equally, queued on a "first come, first served" basis. We will never introduce "paid priority" features in the future — rest assured.

There are no hard usage limits, but to ensure fairness, each IP may have short-term concurrency restrictions during peak hours (e.g., only 1 task can be submitted at a time). If you have batch processing needs, we recommend submitting in batches at different times, or contacting us in advance (see contact info in the footer).

Absolutely not. All audio files you upload are only temporarily stored during processing and are automatically deleted immediately after processing. They are never used for training or sharing. We use encrypted transmission to fully protect your privacy.

Voice Cloning (IndexTTS 2.0): Supports Chinese and English, can clone any speaker's voice.

Voice Synthesis (Qwen3-TTS 1.7B): Supports Chinese, English, Japanese, Korean, and many other languages, with dozens of built-in high-quality voices (male/female, children's voices, emotional styles, etc.).

Check the specific parameters on each tool page.

AI synthesis quality is affected by original audio quality, text punctuation, speech rate, and other factors. We recommend uploading clear audio without background noise, 5-20 seconds long for cloning, and using standard punctuation in your text for better results.

You can submit feedback through the "Contact Us" email at the bottom of the page, and we will reply within 24 hours. We also welcome feature suggestions — we will carefully evaluate and continuously optimize.