English AI Voice

Brian — free American English text to speech

Brian is a free male American English AI voice — a Microsoft neural voice. Type your text below to hear it read aloud, then download the audio. No account needed.

Male American accent Global voice Multilingual Free

Make Brian say your words

Voice details

Language
English
Accent
American
Gender
Male
Type
Global · Microsoft neural
Download
MP3
Speed
Natural pace
Languages
Multilingual
Voice ID
en-US-BrianMultilingualNeural
Price
Free

2 voices are called Brian

This page is the male American English Global voice. Looking for another one?

How Brian sounds — measured

Median pitch

119 Hz

In the middle of 18 male American English voices

Intonation range

5.9 semitones

Steadier than typical for male American English voices (8.9)

Tone (spectral centroid)

2,867 Hz

Brighter, crisper tone than most of its group

Speaking rate

182 words/min

In the middle of 75 English voices measured

Pitch across the sample passage dashed line = Brian’s median

How long your audio will be at 15.7 characters/s

1,000 characters
≈ 1 min 4 s
5,000 characters (free limit per render)
≈ 5 min

Measured by SpeakScribe, October 2026 · How we measure · Every English voice side by side

What Brian fits best

  • Long narration and audiobooks

    Its intonation spans 5.9 semitones against 8.9 for male American English voices — steady enough to stay easy on the ear over many minutes.

Use Brian in a conversation

With Multi-voice, Brian can read one character while Ava (a female English voice) reads another — the whole dialogue comes back as one MP3 with even volume between speakers. A free account can make 3 two-speaker dialogues a month; paid plans allow up to 4 speakers on Starter.

Try a dialogue with Brian →

Formats, limits and help

Brian: quick answers

Measured from its audio, Brian has a median pitch 119 Hz (in the middle of 18 male American English voices); intonation range 5.9 semitones (steadier than typical for male American English voices (8.9)); tone (spectral centroid) 2,867 Hz (brighter, crisper tone than most of its group).
About 1 min 4 s. Brian reads about 15.7 characters a second, spaces included (measured on its sample passage), so the free limit of 5,000 characters per render is about 5 minutes of audio. Numbers, names and long pauses make it longer.