Hume AI
Expressive voice AI that reads and produces emotional tone in speech, with a realtime voice interface.
why this verdict
Downgrade — pay for the workflow, not the AI
ElevenLabs covers the headline feature free, so the $10/mo is now paying for the workflow around it.
- A named launch, not a vibe named, unlinked
- Realtime voice APIs from major labs shipping expressive speech as a standard feature. OpenAI, November 25, 2025 — no announcement link recorded yet.
- How much of the job it covers headline only editorial call
- The headline feature, but not the workflow, data or integrations around it.
- Is there a free way to do it? yes
- 1 of 3 listed replacements have a usable free tier: ElevenLabs.
- What the call is worth $120/yr
- $10/mo at the entry paid tier — $120 a year per seat.
- Threatened by
- OpenAI
- Since
- November 25, 2025
- List price
- $10/mo
- Per year
- $120
The backstory
Hume builds voice models focused on emotional expressiveness, plus an API for measuring feeling in speech. Realtime speech-to-speech from the big labs now handles tone, interruption and laughter natively, which absorbs much of the differentiator. Hume's emotion measurement work still has research and healthcare buyers who need the analysis rather than the voice. The voice product faces much larger incumbents.
Escape hatches
switching could free up $120/yrMarket-leading voice cloning and text to speech
elevenlabs.io open_in_newVery low latency realtime voice models
cartesia.ai open_in_newSpeech recognition and voice agent infrastructure
deepgram.com open_in_new1 of 3 replacements have a usable free tier.