Male · American English
Deep, calm American male — the audiobook narrator voice in the free tier.
curl -X POST /api/v1/audio/speech \
-H "Authorization: Bearer ev_..." \
-d '{"input": "Hello!", "voice": "am_michael"}'Michael is the deeper, slower counterpart to am_adam — a low-bass-baritone with a measured 130–145 wpm cadence, slightly darker timbre, and a quietly authoritative presence that holds up across multi-hour listens. Where Adam is a confident explainer, Michael is the late-night podcast host or the audiobook narrator you trust to take you through a 12-hour nonfiction read without listener fatigue. His delivery is restrained — minimal inflection, clean punctuation discipline, low effort — which is precisely why he scales to long-form so well.
Yes. Michael (am_michael) is on EasyVoice's free tier — 5,000 characters per day, daily reset, no signup or credit card required. That's enough audio for a 4-to-5-minute narrated piece every day, indefinitely. Pro at $9.99/mo flat removes the cap entirely, which matters more for Michael than other voices because audiobook and long-form podcast scripts run far past 5K characters.
Michael is a native American-English (en-US) voice. He is not multilingual — for non-English narration in similar deep-male-narrator territory, EasyVoice's Pro catalog has locale-specific voices (em_alex for Spanish, im_nicola for Italian, hm_omega for Hindi, jm_kumo for Japanese, pm_alex for Portuguese).
Adam is mid-baritone, slightly faster, and more 'explainer' — good for product walkthroughs, training, tech YouTube. Michael is lower, slower, and more 'narrator' — better for audiobooks, sleep audio, and long-form podcasts where you want the listener to settle in for hours. Try both on the same script via EasyVoice's free tier; the right pick is usually obvious within 30 seconds.
Antoni (ElevenLabs) and Mark (PlayHT) sit in the deep-American-male-narrator slot alongside Michael. The realism is comparable in 2026 — modern Kokoro, ElevenLabs and PlayHT models all clear the bar where casual listeners can't tell synthetic from human in a blind A/B. Pricing is the divergence: ElevenLabs Antoni is metered per character on Creator/Pro, PlayHT Mark is similarly usage-based, and EasyVoice am_michael is free up to 5K/day and $9.99/mo unlimited on Pro. For audiobook producers running 50,000-word manuscripts, EasyVoice's flat rate is decisively cheaper.
Yes. Full commercial usage rights are granted on every EasyVoice plan, free tier included. Use Michael for paid audiobooks (Audible, Spotify Audiobooks, Findaway Voices and direct sales), monetized podcasts, paid courses, sleep-app content, and commercial software products. No royalties, no per-project licensing, no attribution required.
Pacing is controlled via punctuation, and Michael's slow baseline cadence means he honours pauses generously — semicolons, em-dashes, and paragraph breaks all produce noticeable breathing room. Full SSML markup is on the roadmap. For audiobook production specifically, well-punctuated source text plus chapter-break paragraph spacing produces excellent results without any markup.