Choosing a voice

AI voices for presentations and e-learning

For presentations and e-learning, choose a clear, bright tone with steady intonation and a moderate pace. Brian has the narrowest intonation range of the seven picks at 5.9 semitones, so his delivery stays level. Rosa has the highest spectral centroid of the seven picks at 3,917 Hz, which gives her tone the most high-frequency energy here.

Recommended voices

Brian

Male · American

GLOBAL

Steady American male narrator

Intonation range 5.9 semitones, narrower than 86% of the 36 male English voices. Median pitch 119.3 Hz at 182 words per minute, near the middle of the English set.

Pitch
119.3 Hz
Range
5.9 st
Pace
182 wpm
Use voice
Rosa

Female · Philippine

GLOBAL

Bright Philippine English voice

Spectral centroid 3,917 Hz, the highest of these seven picks. Median pitch 201.8 Hz and a 7.1-semitone range.

Pitch
201.8 Hz
Range
7.1 st
Pace
157 wpm
Use voice
Ava

Female · American

GLOBAL

Multilingual American female voice

Median pitch 194.4 Hz, a 9.3-semitone range, and 187 words per minute. She is multilingual: the same voice can read other languages.

Pitch
194.4 Hz
Range
9.3 st
Pace
187 wpm
Use voice
Sarah

Female · American

STUDIO

Studio American female voice

Median pitch 191.3 Hz, an 8.9-semitone range, and 191 words per minute. Spectral centroid 2,210 Hz.

Pitch
191.3 Hz
Range
8.9 st
Pace
191 wpm
Use voice
Imani

Female · Tanzanian

GLOBAL

Tanzanian English voice

Median pitch 195.8 Hz and an 8.6-semitone range, at 158 words per minute. Spectral centroid 2,746 Hz.

Pitch
195.8 Hz
Range
8.6 st
Pace
158 wpm
Use voice
Christopher

Male · American

GLOBAL

American male narrator, moderate pace

Median pitch 110.5 Hz, 164 words per minute, and a 10.6-semitone range. Spectral centroid 2,963 Hz.

Pitch
110.5 Hz
Range
10.6 st
Pace
164 wpm
Use voice
Eric

Male · American

GLOBAL

Bright American male narrator

Spectral centroid 3,099 Hz, the brightest of the male picks. Its 11.0-semitone range is the widest of these seven, so it moves more than Brian.

Pitch
106.4 Hz
Range
11.0 st
Pace
161 wpm
Use voice

What makes a voice work for presentations and e-learning

Slides and course screens carry the detail, so the narration must be easy to follow while the learner reads. Three measurements decide that: spectral centroid for clarity, intonation range for steadiness and words per minute for pace.

Clarity: a higher centroid means more high-frequency energy, which sounds crisper. Rosa (3,917 Hz) and Eric (3,099 Hz) are the brightest picks.

Steadiness: a narrow range keeps a long lesson level. Brian's 5.9 semitones is the narrowest of the seven. Eric's 11.0 moves the most, which suits a conversational lecture better than a dense course.

Pace: the picks run from 157 to 191 words per minute. Christopher (164) and Eric (161) are slower, while Sarah (191) is quicker, so leave more full stops between points when using her.

Writing slide and course narration for an AI voice

Write one point per slide line, and let the narration explain what the slide does not show. Reading every bullet word for word wastes the learner's time.

Punctuation controls pauses. A full stop after each point gives the learner a clear beat to look at the slide.

  • Name each point before you explain it.
  • Write numbers, units and acronyms the way the narrator should say them.
  • Use the same term for one concept on every slide.
  • Keep sentences under 20 words, and split steps into numbered sentences.
  • At 160 words per minute, a 30-second slide holds about 80 words.
  • Preview one slide before generating the narration for the whole course.

Male or female voice for presentations?

Both suit the job. The female picks run from 191.3 Hz (Sarah) to 201.8 Hz (Rosa) in median pitch. The male picks run from 106.4 Hz (Eric) to 119.3 Hz (Brian). The lowest female pick sits about 72 Hz above the highest male pick.

Choose by the lesson. A bright female voice like Rosa suits an explainer, and a lower male voice like Christopher suits a training module.

Matching the accent to your learners

This set includes three accents: Philippine English (Rosa), Tanzanian English (Imani) and American English (Brian, Christopher, Eric, Ava and Sarah). Choose the accent your learners already know.

A narrator with a different accent from your learners can still work, but preview two picks on your own script before you choose.

Same criteria in other languages

Hindi

Swara · Female · 255.5 Hz

Swara, a female Indian-accent voice at 255.5 Hz, the highest pitch of the three female Hindi voices measured.

Urdu

Gul · Female · 189.3 Hz

Gul, a female Indian-accent voice at 189.3 Hz with a 5.4-semitone range, the narrowest of these nine voices.

Bengali

Tanishaa · Female · 255.5 Hz

Tanishaa, a female Indian-accent voice at 255.5 Hz with a 12.8-semitone range.

Arabic

Hamdan · Male · 130.5 Hz

Hamdan, a male Emirati voice at 130.5 Hz with a 15.6-semitone range, the widest of these nine voices.

Spanish

Maria · Female · 189.3 Hz

Maria, a female Costa Rican voice at 189.3 Hz with a 7.6-semitone range.

Persian

Dilara · Female · 183.1 Hz

Dilara, a female standard Iranian Persian voice at 183.1 Hz with a 9.1-semitone range, suited to explainers and educational videos.

Tamil

Pallavi · Female · 241.5 Hz

Pallavi, a female Indian voice at 241.5 Hz with a 10.4-semitone range.

Portuguese

Alex · Male · 132.9 Hz

Alex, a male Brazilian Studio voice at 132.9 Hz with an 8.4-semitone range.

French

Fabrice · Male · 127.3 Hz

Fabrice, a male Swiss voice at 127.3 Hz with a 12.2-semitone range, the second-highest pitch of the seven male French voices measured.

Frequently asked questions

For steady narration, Brian (5.9 semitones). For a clear, bright tone, Rosa (3,917 Hz). Preview both on one slide before you choose.
Generate one audio clip per slide, then insert each clip on its slide in PowerPoint. Write each slide's script as one point per line, and preview one slide before generating the rest.
Christopher, a male American voice at 164 words per minute, or Imani, a Tanzanian English voice with an 8.6-semitone range.
Yes, for the languages in our library. Gul (Urdu, 189.3 Hz), Swara (Hindi, 255.5 Hz) and Dilara (Persian, 183.1 Hz) are examples.
Choose by the measured numbers: steady intonation and a moderate pace. Brian (5.9 semitones, 182 words per minute) and Ava (187 words per minute) are both multilingual voices.
Rosa for a bright, clear tone at 3,917 Hz. Sarah, a Studio American voice at 191 words per minute, for a faster delivery.

Try these voices on your own script

Paste your text in Voice Studio, pick a voice from this page and preview it before you generate.

Open Voice Studio