Chirp 3 HD
Google's Gemini-powered HD text-to-speech — 30 voices, 50+ languages
About
Chirp 3 HD is Google Cloud's late-2024 flagship TTS voice family, built on Gemini-era generative models. It offers 30 HD voices (14 female, 16 male) with natural prosody that community migrators consistently describe as a large step up from WaveNet and Neural2. The API supports 50+ languages at GA with steady expansion through 2025, plus streaming synthesis, batch synthesis, pace and pause control, custom pronunciations (IPA/X-SAMPA), and SSML in preview. Pricing follows Google Cloud's per-character model with a 1M character monthly free quota. Limits: voice choice is fixed presets (no cloning, no voice design); quality varies by language; and some locales (e.g. Traditional Chinese at launch) arrived later than others.
Specs
Context window
Pricing
Free quota
API access
Weights
Released
Benchmarks
Community-reported: substantially more natural prosody than WaveNet
Provider-reported unless a source says otherwise. Benchmarks are marketing-adjacent by nature — treat them as directional.
Use cases
Trust score — 8/10
Scored 8/10 (verified tier, metrics-v1). Evidence: Google Cloud release notes document 30 HD voices, 50+ languages GA, expanding coverage through 2025; Multiple independent GitHub projects migrated to Chirp 3 HD citing large prosody gains over WaveNet; Gemini-based generative TTS; 1M chars/month free quota corroborated.