← Voice & audio

Whisper

OpenAI's open speech-to-text model, runs locally.

Pricing
Free, open source
Free tier
Yes
Open source
Yes
Score
93/100 (public evidence)

Evidence score: 93/100

Scored from public evidence (hands-on testing coming). Evidence gathered 2026-10-16; 100/100 points verifiable. Unverifiable criteria are excluded, not guessed.

CriterionEvidencePointsSource
Free tiertrue20/20directory data verified 2026-10-16
Pricing clarityfree, open source (no pricing needed)15/15link
Docs qualityREADME/docs in repo20/20link
Platform supportWindows, macOS, Linux15/15link
Open sourcetrue15/15link
Recent updateslast commit pushed 45 days ago8/15link

Hands-on test notes: coming. Pricing checked 2026-10-16; prices change, so confirm on the vendor site.

Visit Whisper ↗

Alternatives

ElevenLabs

Realistic text-to-speech and voice cloning.

Free tier 75/100 · evidence
Suno

Generate full songs from a prompt.

Free tier 82/100 · evidence
Murf AI

Studio-quality AI voiceovers.

Free tier 76/100 · evidence