A voice file, a reading voice, or a cleaner take
AI voice and TTS tools
Use this page when the output is speech: a synthesized voice, a reading of a page, a dub, or a cleaned recording. A good fit is a listing that says so. A music generator is a different page in the index, and it is not repeated here.
ElevenLabs, MiniMax, Murf, and PlayHT generate or clone voices. Speechify and NaturalReader read text aloud. Descript edits media like a document and includes overdub. Adobe Podcast cleans speech. 喜马拉雅 is the mainland podcast and audiobook listing. 讯飞星火 is a chat assistant whose summary includes speech.
MiniMax, 喜马拉雅, and 讯飞星火 are marked offered in mainland China. The others are not confirmed. Voice-consent rules are not a field in the index.
How to choose
- Choose ElevenLabs, Murf, or PlayHT when you need a generated voice or a dub and an unconfirmed China mark is acceptable. Choose MiniMax when that voice product should be marked offered in mainland China.
- Choose Speechify or NaturalReader when the job is reading a document or a page, not performing a script as a character.
- Choose Descript when you will edit the recording like a transcript. Choose Adobe Podcast when the job is cleaning speech, including the free Enhance Speech the listing names.
- Choose 喜马拉雅 for audiobooks and podcasts aimed at mainland audiences. Choose 讯飞星火 when speech sits inside the same assistant as chat and writing. For a talking-head video, use the HeyGen comparison.
Shortlist
Ten audio and speech listings. Music generators such as Suno stay in the audio category and off this shortlist.
AudioFree tier
ElevenLabs
Speech synthesis, voice cloning, and dubbing, with a free tier and paid usage.
Mainland China: not confirmed
View listingOpen this for speech synthesis, voice cloning, and dubbing, with a free tier and paid usage. Mainland China access is not confirmed.
AudioFree tier
MiniMax Audio
MiniMax 语音
MiniMax's speech synthesis and voice products listed on its public mainland site.
Mainland China: offered
View listingOpen this for MiniMax speech synthesis and voice products on the public mainland site. Marked offered in mainland China, and freemium in the index.
AudioFree tier
Murf
Studio for AI voiceovers and dubbing, with free trials and paid creator plans.
Mainland China: not confirmed
View listingOpen this for a studio of voiceovers and dubbing, with free trials and paid creator plans. Mainland China access is not confirmed.
AudioFree tier
PlayHT
AI voice generation API and studio for clones and multilingual narration.
Mainland China: not confirmed
View listingOpen this for a voice API and studio aimed at clones and multilingual narration. Mainland China access is not confirmed.
AudioFree tier
Speechify
Text-to-speech reader for documents and web pages, with free voices and Premium.
Mainland China: not confirmed
View listingOpen this to read documents and web pages aloud, with free voices and Premium. Mainland China access is not confirmed.
AudioFree tier
NaturalReader
TTS reader for students and professionals, free online voices and paid desktop plans.
Mainland China: not confirmed
View listingOpen this for a TTS reader aimed at students and professionals, with free online voices and paid desktop plans. Mainland China access is not confirmed.
AudioFree tier
Descript
Audio and video editor that edits media like a document, with AI overdub and free trial plans.
Mainland China: not confirmed
View listingOpen this when you edit audio or video like a document and want overdub. The listing includes free trial plans. Mainland China access is not confirmed.
AudioFree tier
Adobe Podcast
Adobe's web tools for cleaning speech audio, with free Enhance Speech on the public site.
Mainland China: not confirmed
View listingOpen this for Adobe’s web tools that clean speech audio, with free Enhance Speech on the public site. Mainland China access is not confirmed.
AudioFree tier
Ximalaya AI Studio
喜马拉雅创作中心
Ximalaya's creator tools for audiobooks and podcasts aimed at mainland audiences.
Mainland China: offered
View listingOpen this for Ximalaya’s creator tools for audiobooks and podcasts aimed at mainland audiences. Marked offered in mainland China.
ChatFree tier
Spark
iFlytek's assistant for chat, writing, and speech-related tasks.
Mainland China: offered
View listingOpen this for iFlytek’s assistant when chat, writing, and speech belong together. Marked offered in mainland China. It is not a TTS studio.
Head to head
Related pages
Questions
Which listings are text-to-speech?
ElevenLabs, MiniMax, Murf, and PlayHT describe synthesis, cloning, dubbing, or narration. Speechify and NaturalReader describe reading documents or pages aloud. Descript is an editor with overdub. Adobe Podcast cleans speech. 喜马拉雅 is audiobooks and podcasts. 讯飞星火 is an assistant that includes speech.
What is offered in mainland China?
MiniMax Audio, 喜马拉雅, and 讯飞星火 are marked offered. ElevenLabs, Murf, PlayHT, Speechify, NaturalReader, Descript, and Adobe Podcast are not confirmed. The mark is a public offer from the seed, not a connection test.
Where is music?
Suno, Udio, AIVA, Boomy, and Soundraw are music listings in the audio category. Their summaries write songs or soundtracks, not a speaking voice. They stay off this shortlist. Browse Audio if that is the job.
Does a free tier let me clone a voice?
The index does not say that. ElevenLabs and PlayHT mention cloning in the summary and are freemium. Consent, verification, and what the free tier includes are on the vendor’s site. This page does not describe a way around those rules.
Are these free?
All ten are freemium in the index. Adobe Podcast’s listing names free Enhance Speech. Speechify names free voices. The index does not copy character caps or commercial-license terms.
Limits
NavBoth is a directory of public products already in the index. These pages help you choose a listing. The product itself runs on the vendor’s site. Mainland China marks come from public offers reviewed with the seed; they are not a live connection test. No listing on these pages has a paid badge. Names belong to their owners.
Commercial use of a cloned voice, watermarking, and language coverage stay on the vendor’s site. This page records the speech job each listing already describes.