贴心女声/Candy
"贴心女声/Candy" is an official ListenHub Mandarin Chinese female voice, suited to video dubbing, with a natural delivery.
ListenHub · Voice Library · 12 languages
ListenHub speaks with 909 official voices across 12 languages, and every one of them plays right here. Find the one that fits, then take it straight into Text to Speech.
204 of 491 voices
Clear filters"贴心女声/Candy" is an official ListenHub Mandarin Chinese female voice, suited to video dubbing, with a natural delivery.
"古风少御" is an official ListenHub Mandarin Chinese female voice, suited to audiobooks, with a natural delivery.
"温柔淑女" is an official ListenHub Mandarin Chinese female voice, suited to audiobooks, with a natural delivery.
"Vivi 2.0" is an official ListenHub Mandarin Chinese female voice, suited to general audio creation, with a natural and neutral delivery.
"清脆少女" is an official ListenHub Mandarin Chinese female voice, suited to general audio creation, with a emotional delivery.
"傲娇御姐" is an official ListenHub Mandarin Chinese female voice, suited to general audio creation, with a natural and neutral delivery.
"软软女孩" is an official ListenHub Mandarin Chinese female voice, suited to general audio creation, with a emotional delivery.
"甜美女声" is an official ListenHub Mandarin Chinese female voice, suited to general audio creation, with a emotional and cute and playful delivery.
"温暖少女" is an official ListenHub Mandarin Chinese female voice, suited to general audio creation, with a emotional delivery.
"淡雅学姐" is an official ListenHub Mandarin Chinese female voice, suited to general audio creation, with a natural and neutral delivery.
"嗲嗲学妹" is an official ListenHub Mandarin Chinese female voice, suited to general audio creation, with a natural and neutral delivery.
"成熟女性音色" is an official ListenHub Mandarin Chinese female voice, suited to general audio creation, with a natural and neutral delivery.
"成熟女性音色-beta" is an official ListenHub Mandarin Chinese female voice, suited to general audio creation, with a natural and neutral delivery.
"少女音色" is an official ListenHub Mandarin Chinese female voice, suited to general audio creation, with a natural and neutral delivery.
"少女音色-beta" is an official ListenHub Mandarin Chinese female voice, suited to general audio creation, with a natural and neutral delivery.
"甜美女性音色" is an official ListenHub Mandarin Chinese female voice, suited to general audio creation, with a cute and playful delivery.
"甜美女性音色-beta" is an official ListenHub Mandarin Chinese female voice, suited to general audio creation, with a cute and playful delivery.
"御姐音色" is an official ListenHub Mandarin Chinese female voice, suited to general audio creation, with a natural and neutral delivery.
"御姐音色-beta" is an official ListenHub Mandarin Chinese female voice, suited to general audio creation, with a natural and neutral delivery.
"萌萌女童" is an official ListenHub Mandarin Chinese female voice, suited to children's content, with a natural and neutral delivery.
"俏皮萌妹" is an official ListenHub Mandarin Chinese female voice, suited to general audio creation, with a cute and playful delivery.
"甜心小玲" is an official ListenHub Mandarin Chinese female voice, suited to general audio creation, with a natural and neutral delivery.
"妩媚御姐" is an official ListenHub Mandarin Chinese female voice, suited to general audio creation, with a natural and neutral delivery.
A high-pitched, childish voice with a whiny, pouty quality at a moderately fast pace, full of coquettish and complaining emotions, suitable for animated characters or expressing delicate dissatisfaction.
Each of the 12 languages has its own voices: the Chinese set is not the English one translated, it is 491 different voices. Pick the language you will publish in first.
Scene tags say what a voice is for: podcast, audiobook, documentary, kids. Style tags say how it delivers. Stack two of them and a few hundred voices become a handful.
Every card plays the same demo line, so what you are comparing is the voice rather than the script it happens to be reading.
Read text aloud opens Text to Speech with that voice already selected. The existing three-language voices also keep their Podcast and AI Voice shortcuts.
Every voice here has a speaker ID, and that is what ListenHub Skills, the MCP server, the CLI and the API all use to pick a voice. Copy one off a card, paste it to your assistant, and the audio it makes comes back in that voice.
One command inside your project. Claude Code, Cursor, Windsurf and OpenCode all read Agent Skills.
npx skills add marswaveai/skillsCreate a key starting with lh_sk_. Your assistant asks for it the first time it needs one and keeps it after that.
Create an API keyPress Copy agent prompt on any card above. The sentence arrives with the voice name and its speaker ID already in it — add whatever you want read, and send.
Use the ListenHub voice "贴心女声/Candy" (speakerId: doubao-official-zh_female_tiexinnvsheng_mars_bigtts) to turn the following into audio:Any card gives you the same sentence with its own voice in it.
Duo episodes need two voices that sound like different people, not the same person twice. Play them side by side before you commit to the pair.
An audiobook or a long read lives or dies on whether the voice is still pleasant forty minutes in. Steady beats striking here.
Teaching material wants clear over characterful — the documentary and education tags are where those voices sit.
Video dubbing and character work need range rather than neutrality. Filter by the character and multi-emotion styles to find voices that act.
No. Every demo on this page plays without signing in. An account only comes into it when you want to generate audio of your own.
909 today across 12 languages, including 142 English, 491 Chinese and 35 Japanese voices. The page counts what is actually published rather than quoting a stale hardcode.
Listening is free. Generating audio spends credits from your plan, and what it costs depends on the format and length. A few voices are marked Subscription: you can audition them here, but generating with them needs a plan.
Scene is what a voice is for — podcast, audiobook, news, meditation. Style is how it speaks — neutral, conversational, character, multi-emotion. They come from the same vocabulary the voice picker uses inside the app, so a voice you find here behaves the way you expect once you are in the editor.
Yes. A duo podcast is written for two hosts and you choose both of them, which is exactly why it helps to audition them next to each other first.
Yes, through voice cloning — 25 to 35 seconds of clean audio and your voice joins your own picker. Cloning supports the same 12 languages as Text to Speech. Cloned voices stay private to your account and never appear in this public library.
Every published profile carries scene or style tags and a written description. Use the tags to narrow the list, read the description for intent, then press play to judge the voice itself.
When you want two hosts talking it through rather than one voice reading it out.
When the script is already written and you just need it spoken.
When the delivery matters — several voices in one take, or a line that has to land with real emotion.
When none of the library voices should be the one speaking — record 30 seconds and use your own.
Every voice, id and tag in one plain-text fetch, with no pages to walk.
https://listenhub.ai/voices.txt