Best for Speech & Voice

All models2292Chat & Knowledge558Coding381Agents & Tool Use541Writing & Language360Vision & Documents395Image Generation & Editing183Video Generation148Speech & Voice358Embeddings & Retrieval671Decision models118Music & Sound29
Text to speech136Speech to text151Realtime55Translation8Voice changers4
ModelEVA-Bench experienceLanguagesPrice coversPriceReleased
Gemini 3.8 LiveGoogle
88.6
en, es, fr +97Provider-stated input/output audio minute components; text, tools and session/backend charges remain separate.$0.005 / $0.018 / minuteSep 2026
Gemini 3.8 Live Extended ThinkingGoogle
79.2minimal
en, es, fr +97Provider-stated input/output audio minute components; text, tools and session/backend charges remain separate.$0.005 / $0.018 / minuteSep 2026
Grok Voice 3.0xAI
71.2
en, es-MX, es-ES +17—$0.08 / minuteOct 2026
Grok Voice 2.0xAI
64.1unstated
—session audio; text input charged separately$0.08 / minute
GPT Realtime 2.1OpenAI
61.6unstated
——$32 / $64 / 1M audio tokensJul 2026
Gemini 3.1 Flash Live PreviewGoogle
58.9unstated
en, es, fr +97Provider-stated input/output audio minute components; text, tools and session/backend charges remain separate.$0.005 / $0.018 / minuteMar 2026
GPT Realtime 1.5OpenAI
56.6
——$32 / $64 / 1M audio tokensFeb 2026
GPT Realtime MiniOpenAI
40.6
——$10 / $20 / 1M audio tokensOct 2025
GPT Realtime 2OpenAI
39unstated
——$32 / $64 / 1M audio tokensMay 2026
GPT RealtimeOpenAI
36.5
——$32 / $64 / 1M audio tokensAug 2025
GPT Realtime 2.1 MiniOpenAI
29.2unstated
——$10 / $20 / 1M audio tokensJul 2026
UUltravox RealtimeUltravoxOpen
2.9
en, es, fr +38—$0.05 / minuteDec 2025