←Back to NewsAI News/AudionewsAudioBenchmarksArtificial Analysis' Benchmark Reveals Gemini 3.1 Flash TTS Beats Sonic 3.6 on PronunciationA new benchmark scores text to speech models on how correctly they pronounce tricky words, and voice preference does not predict accuracy.SourceArtificial AnalysisPublishedSep 22, 2026, 2:26 AMAuthorAlphaSignal NewsroomRead1 min readA new benchmark scores text to speech models on how correctly they pronounce tricky words, and voice preference does not predict accuracy.Reporting is indexed from AlphaSignal. Rights remain with the original publisher and cited sources.Read original report ↗Next readsArtificial Analysis · newsMicrosoft's MAI-Transcribe-2 Beats ElevenLabs on Accuracy at 54% Lower PriceAI at Meta · newsMeta's Muse Voice Transcribe Beats Every Rival With 3.1% Error RateSesame · newsSesame's TurnBench Exposes How Gemini Live and OpenAI Realtime Fumble Conversations
