voices.catalog The voice library.
The voice library.
Forty-two of them.
Not stock TTS. Each voice is built for a kind of piece. The News Anchor moves fast because news wants that. The Scholar breathes because essays need the breath. Pick one and hear a sample.
VOICES 42
LANGUAGES 6 (→ 26)
SAMPLE RATE 48 kHz
BITRATE 256 kbps
FORMATS MP3 · AAC · OPUS
PHRASING MODEL v2.4 · april.26
// calm
The Narrator
Your default. Unhurried, warm, present. The voice most pieces want.
EN · FR · ES · DE · IT · PT
146 wpm / 172 Hz
The Scholar
Thoughtful cadence. Pauses where you'd pause reading. For essays.
EN · FR · DE
134 wpm / 118 Hz
Late Night
Lower register. Slower breath. For the 2am read.
EN · FR · ES
122 wpm / 86 Hz
Sunday Morning
Soft. Unhurried. Coffee still hot.
EN · IT · PT
132 wpm / 156 Hz
// sharp
News Anchor
Crisp. Declarative. Tight pacing. For news and analysis.
EN · FR · DE · ES
172 wpm / 108 Hz
Market Desk
Fast. Clean. Numbers read right. For finance and tech.
EN · DE
182 wpm / 132 Hz
Editorial
Measured but opinionated. For op-eds and reviews.
EN · FR · IT
150 wpm / 124 Hz
// human
Your Friend
Warm, casual. Like your smart friend explaining the piece.
EN · FR · ES · PT
158 wpm / 128 Hz
The Tinkerer
Curious, a little nerdy. For how-tos and technical pieces.
EN · DE
164 wpm / 145 Hz
Field Reporter
A touch of gravel. Sounds like they've been there. For long-form.
EN · ES
144 wpm / 98 Hz
// signature
Documentary UK
British. Unhurried. A bit dry. For essays and history pieces.
EN
140 wpm / 95 Hz
The Curator
French-accented English. For art, design, culture.
EN · FR
138 wpm / 162 Hz
West Coast
Californian, a little laid-back. For tech and lifestyle.
EN
154 wpm / 138 Hz
Brooklyn Lit
New York literary. Brisk, a little ironic. For criticism.
EN
168 wpm / 142 Hz
custom
Want your own voice in there?
Studio tier lets you train a private voice from twenty minutes of audio. Your own voice, your colleague's, a public-domain actor. It stays yours. Nobody else can use it.
see Studio tier===============================================================================
listen_first ?