English text to speech voices

Which English voice should you choose?

• For videos, audiobooks and ads: start with the 6 HD voices, such as Adam HD and Elliot HD. They have the richest, most natural sound.
• For long documents, lessons and phone messages: the 43 standard voices, such as Natasha and William, stay clear and consistent over long texts.
• For a specific character: the 80 styled voices are each made for one job, from explainer, luxury brand, podcast, commercial and audiobook. Try Audrey and Brandon.
• For casual, conversational content: the 36 natural voices, such as Alice and Daniel Hughes, talk rather than read.
• For a local audience: pick a voice from your listeners' country. English voices here cover 14 accents.

English accents

Pronunciation, rhythm and even some words change from country to country. These are the accents available:

• United States (70 voices): General American English, the accent most viewers know from films, TV and YouTube. It is the safe choice for a global audience.
• United Kingdom (29 voices): British English with Received Pronunciation features: non-rhotic "r", crisp consonants and a polished, trusted sound.
• Australia (11 voices): Australian English, friendly and relaxed, with its typical vowel shifts and rising intonation.
• Canada (11 voices): Canadian English, close to American English but with its own vowels, a natural fit for Canadian brands.
• India (11 voices): Indian English, clear and syllable-timed, familiar to hundreds of millions of listeners in India and South Asia.
• Ireland (10 voices): Irish English, warm and melodic, with a soft "t" and a lilting rhythm.
• South Africa (9 voices): South African English, clear and confident, with influences from Afrikaans and other local languages.
• Hong Kong (2 voices): Hong Kong English, clear and precise, influenced by Cantonese rhythm.
• Kenya (2 voices): Kenyan English, clear and even, widely used in East African media and business.
• New Zealand (2 voices): New Zealand English, close to Australian English but with its own distinctive vowels.
• Nigeria (2 voices): Nigerian English, rhythmic and expressive, spoken by one of the largest English-speaking populations in the world.
• Philippines (2 voices): Philippine English, clear and neutral, familiar from customer service and education worldwide.
• Singapore (2 voices): Singapore English, crisp and quick, used in business across South East Asia.
• Tanzania (2 voices): Tanzanian English, measured and clear, often heard alongside Swahili.

How to write English text that sounds natural

• Spell out abbreviations that can be read two ways, for example "Doctor" instead of "Dr." when it starts a sentence.
• Write years and money the way you want them spoken: "twenty twenty-six" or "2026", "$5" or "five dollars".
• Use commas and full stops to control breathing; a long sentence without punctuation is read in one breath.
• Pick the accent your audience expects: American English for most YouTube content, British or Australian for local audiences.

More English AI tools

• Voice Cloning in English: clone a voice from a short sample and make it speak.
• Voice Changer in English: turn a recording into another voice.
• Voice Design in English: create a brand new voice from a written description.
• Speech to Text in English: turn audio into text with sentence timestamps.
• AI Dubbing in English: translate and re-voice a video.