AI Voice Library

Browse, preview, and compare 100+ AI voices across 24+ models. Find the perfect voice for your project.

101+ Voices

101 voices found

--
标准标准标准标准 Chinese Female
标准标准标准标准 Chinese Male
标准标准标准标准 English Female
标准标准标准标准 English Female
标准标准标准标准 English Female
标准标准标准标准 English Female
标准标准标准标准 English Male
标准标准标准标准 English Male
标准标准标准标准 English Male
标准标准标准标准 English Male
标准标准标准标准 English Male
标准标准标准标准 English Male
标准标准标准标准 French Female
标准标准标准标准 French Male
标准标准标准标准 German Female
标准标准标准标准 German Male
标准标准标准标准 Hindi Male
标准标准标准标准 Italian Male
标准标准标准标准 Japanese Female
标准标准标准标准 Japanese Male
标准标准标准标准 Korean Female
标准标准标准标准 Korean Male
标准标准标准标准 Polish Male
标准标准标准标准 Portuguese Male
标准标准标准标准 Russian Male
标准标准标准标准 Spanish Female
标准标准标准标准 Spanish Male
标准标准标准标准 Turkish Male
Premium English Neutral
标准标准标准标准 Chinese Female
标准标准标准标准 Chinese Male
标准标准标准标准 English Female
标准标准标准标准 English Male
标准标准标准标准 Japanese Female
标准标准标准标准 English Neutral
标准标准标准标准 English Neutral
标准标准标准标准 Chinese Neutral
标准标准标准标准 English Neutral
自由 English Male
自由 Portuguese Male
自由 Spanish Male
自由 Hindi Female
自由 Japanese Female
自由 English Female
自由 Spanish Female
自由 Portuguese Female
自由 English Female
自由 English Male
自由 Japanese Female
自由 English Female
自由 English Female
自由 English Male
自由 English Male
自由 Italian Male
自由 English Female
自由 Hindi Male
自由 Italian Female
自由 English Female
自由 French Female
自由 English Female
自由 Chinese Female
自由 Chinese Female
自由 Chinese Female
自由 Chinese Male
自由 Chinese Female
自由 English Female
自由 English Female
自由 French Female
自由 Japanese Female
自由 Korean Female
自由 Spanish Female
Premium English Neutral
标准标准标准标准 English Male
标准标准标准标准 English Female
标准标准标准标准 English Female
标准标准标准标准 English Male
标准标准标准标准 English Female
标准标准标准标准 English Female
标准标准标准标准 English Male
标准标准标准标准 English Female
标准标准标准标准 English Neutral
自由 English Male
自由 English Female
自由 English Female
自由 English Female
自由 English Male
自由 English Male
自由 English Male
标准标准标准标准 English Male
标准标准标准标准 English Male
标准标准标准标准 English Male
标准标准标准标准 Japanese Female
标准标准标准标准 English Male
标准标准标准标准 English Female
标准标准标准标准 Korean Female
标准标准标准标准 Chinese Male
标准标准标准标准 English Female
标准标准标准标准 English Neutral
Premium English Neutral
Premium English Neutral
自由 English Neutral

声音不匹配过滤器。 尝试调整搜索条件 。

Voices by AI Model

Each TTS model has its own set of voices with unique characteristics. Some models support voice cloning, allowing you to use any voice as a reference.

BarkBark 28 voices 标准标准标准标准

尝试模型

Transformer-based text-to-audio model that generates realistic speech, music, and sound effects.

Chinese Female 1

Chinese
Use

Chinese Male 1

Chinese
Use

English Female 1

English
Use

English Female 2

English
Use

English Female 3

English
Use

English Female 4

English
Use

English Male 1

English
Use

English Male 2

English
Use

View all 28 Bark voices

ChatterboxChatterbox 1 voices Premium

尝试模型

State-of-the-art zero-shot voice cloning with emotion control from Resemble AI.

Default

English
Use

CosyVoice 2CosyVoice 2 5 voices 标准标准标准标准

尝试模型

Alibaba's scalable streaming TTS with human-parity naturalness and near-zero latency.

Chinese Female

Chinese
Use

Chinese Male

Chinese
Use

English Female

English
Use

English Male

English
Use

Japanese Female

Japanese
Use

Dia TTSDia TTS 2 voices 标准标准标准标准

尝试模型

Multi-speaker dialog generation model that creates natural conversations between speakers.

Speaker 1

English
Use

Speaker 2

English
Use

GPT-SoVITSGPT-SoVITS 1 voices 标准标准标准标准

尝试模型

Few-shot voice cloning TTS that replicates any voice from just 5 seconds of audio.

Default

Chinese
Use

IndexTTS-2IndexTTS-2 1 voices 标准标准标准标准

尝试模型

Zero-shot TTS with fine-grained emotion control and high expressiveness.

Default

English
Use

KokoroKokoro 26 voices 自由

尝试模型

Lightweight 82M parameter model delivering studio-quality speech with blazing-fast inference.

Adam

English
Use

Alex

Portuguese
Use

Alex

Spanish
Use

Alpha

Hindi
Use

Alpha

Japanese
Use

Bella

English
Use

Dora

Spanish
Use

Dora

Portuguese
Use

View all 26 Kokoro voices

MeloTTSMeloTTS 7 voices 自由

尝试模型

High-quality multilingual text-to-speech that runs on CPU with minimal latency.

Chinese

Chinese
Use

English British

English
Use

English US

English
Use

French

French
Use

Japanese

Japanese
Use

Korean

Korean
Use

Spanish

Spanish
Use

OpenVoiceOpenVoice 1 voices Premium

尝试模型

Instant voice cloning with granular control over style, emotion, and accent.

Default

English
Use

OrpheusOrpheus 8 voices 标准标准标准标准

尝试模型

Human-level emotional TTS model trained on 100K hours of speech data.

Dan

English
Use

Jess

English
Use

Leah

English
Use

Leo

English
Use

Mia

English
Use

Tara

English
Use

Zac

English
Use

Zoe

English
Use

Parler TTSParler TTS 1 voices 标准标准标准标准

尝试模型

Describe the voice you want in natural language and Parler generates matching speech.

Default

English
Use

PiperPiper 7 voices 自由

尝试模型

A fast, local neural text to speech system optimized for Raspberry Pi and embedded devices.

Alan (UK)

English
Use

Alba (UK)

English
Use

Amy (US)

English
Use

Jenny (UK)

English
Use

Joe (US)

English
Use

Lessac (US)

English
Use

Ryan (US)

English
Use

Qwen3 TTSQwen3 TTS 9 voices 标准标准标准标准

尝试模型

Alibaba's multilingual TTS with voice cloning, preset voices, and voice design from text.

Aiden

English
Use

Dylan

English
Use

Eric

English
Use

Ono Anna

Japanese
Use

Ryan

English
Use

Serena

English
Use

Sohee

Korean
Use

Uncle Fu

Chinese
Use

View all 9 Qwen3 TTS voices

Spark TTSSpark TTS 1 voices 标准标准标准标准

尝试模型

Voice cloning TTS with controllable emotion and speaking style via prompts.

Default

English
Use

StyleTTS 2StyleTTS 2 1 voices Premium

尝试模型

Human-level text-to-speech through style diffusion and adversarial training.

Default

English
Use

Tortoise TTSTortoise TTS 1 voices Premium

尝试模型

Multi-voice text-to-speech focused on quality with autoregressive architecture.

Random

English
Use

VITSVITS 1 voices 自由

尝试模型

Conditional variational autoencoder with adversarial learning for end-to-end text-to-speech.

Default

English
Use

理解大赦国际的呼声

Voice Quality Tiers

TTS.ai offers voices across three quality tiers. Free-tier voices from Piper, VITS, and MeloTTS deliver fast, good-quality synthesis at no cost. Standard-tier voices from models like Kokoro and CosyVoice 2 offer more natural prosody and emotion. Premium-tier voices from OpenVoice, Chatterbox, and StyleTTS 2 provide the most realistic, human-like speech available in open-source TTS.

多语言之声

Many voices support multiple languages. Some models like CosyVoice 2 and GPT-SoVITS support cross-lingual synthesis, where a voice trained in one language can speak naturally in another. The language filter above lets you find voices that natively support your target language, ensuring the best pronunciation and intonation.

Voice Cloning

Some models support voice cloning, which means you can use any voice as a reference to create speech that sounds like that person. Upload a short audio sample (10-30 seconds) and the model will adapt to match the voice characteristics. Models that support cloning include GPT-SoVITS, CosyVoice 2, and Chatterbox.

选择右声

最佳声音取决于您的使用案例。 对于音频书籍和播客, 使用自然流传的优异声音。 对于游戏字符, 探索不同模式的不同声音。 对于无障碍和屏幕阅读器, 选择清晰、 速度快的声音。 对于快速原型, 自由级的声音可以提供即时结果, 无需信用成本 。 在做出选择之前, 用播放按钮预览每个声音 。

常问问题

TTS.ai offers 100+ AI voices across 24 text-to-speech models. Voices span multiple languages, genders, accents, and speaking styles. New voices are added regularly as we expand our model library.

Yes, many voices have audio previews you can listen to directly on this page. Click the play button next to any voice with a preview to hear a sample. You can also test any voice on the Text to Speech page with your own text.

Use the filter controls at the top of the page to narrow voices by model, language, or gender. You can combine filters to find exactly the voice you need — for example, female English voices from the Kokoro model.

Free voices (Kokoro, Piper, VITS, MeloTTS) require no credits. Standard voices (Bark, CosyVoice 2, Dia, Fish Speech) cost 2 credits per 1K characters. Premium voices (Chatterbox, Tortoise) cost 4 credits per 1K characters and offer the highest quality.

Kokoro (free tier) is rated 5/5 for quality and is the most natural-sounding free option. For premium quality, Chatterbox and Tortoise offer exceptional naturalness with voice cloning support. Listen to the previews to judge which voice suits your needs best.

Yes, all voices can be used commercially. Our models use open-source licenses (MIT, Apache 2.0). The generated audio is yours to use in videos, podcasts, apps, games, and any other commercial application.

Our voice library covers 30+ languages including English, Spanish, French, German, Italian, Portuguese, Chinese, Japanese, Korean, Arabic, Russian, Hindi, Dutch, Polish, Turkish, and many more. Language availability varies by model.

Yes, use our Voice Cloning tool to create a custom voice from just 5-30 seconds of reference audio. Cloned voices appear in your account under "My Voices" and can be reused for future text-to-speech generations.

Consider your use case: for audiobooks, choose expressive voices like those from Bark or Chatterbox. For apps and IVR, choose clear voices from Kokoro or MeloTTS. For multilingual content, use CosyVoice 2 or GPT-SoVITS. Preview several options to find the best fit.

Yes, several models offer accent varieties. MeloTTS provides American, British, Indian, and Australian English accents. Other models have regional voice variants for Spanish, French, Portuguese, and Chinese. Filter by language to explore accent options.

Yes, all voices are accessible through our REST API. Specify the model and voice ID in your API request to generate speech with any voice programmatically. See our API Documentation page for code examples and voice ID references.

We regularly add new voices as we integrate additional TTS models and expand existing ones. Follow our updates to stay informed about new voice additions, model improvements, and language expansions.

记录、增强和变声

Preview any voice, then use it directly in Text to Speech. Sign up free and get 50 credits to try premium voices.