音声対話AIモデル&プロバイダーのリーダーボード

音声対話モデルとAPIプロバイダーの分析・比較です。Artificial Analysisは、推論品質、会話ダイナミクス、生成時間、料金など、さまざまな特性に基づいて音声対話モデルとホスティングプロバイダーを分析しています。

詳細は方法論ページをご覧ください。

Weighted average of Speech Reasoning, Agentic Performance, Arena Preference, and Task Success Rate results · Higher is better
Time to first audio (seconds) on Big Bench Audio · Lower is better
Cost to complete fixed 40-question Big Bench Audio subset based on length of input audio, normalized to hourly basis · Lower is better

Artificial Analysis Speech to Speech Index

Artificial Analysis Speech to Speech Index

Weighted average of Speech Reasoning, Agentic Performance, Arena Preference, and Task Success Rate results · Only models with all data available shown · Higher is better

音声推論

音声推論(Big Bench Audio)

Speech reasoning: based on the Artificial Analysis Big Bench Audio dataset · Higher is better

エージェント性能

エージェント性能(𝜏-Voice)

Proportion of replica customer service scenarios resolved while acting as a customer support agent, based on the 𝜏-Voice benchmark · Higher is better · Only full duplex models

注:1回の試行に基づくモデル:GPT-Live-1 (Astra, medium), OpenAI, GPT-Live-1 (Sol, low), OpenAI, GPT-Realtime-2 (Minimal), OpenAI; 2回の試行に基づくモデル:GPT-Realtime-2.1 High, OpenAI

ドメイン別エージェント性能(𝜏-Voice)

Proportion of replica customer service scenarios resolved by domain, based on the 𝜏-Voice benchmark · Higher is better · Only full duplex models

注:1回の試行に基づくモデル:GPT-Live-1 (Astra, medium), OpenAI, GPT-Live-1 (Sol, low), OpenAI, GPT-Realtime-2 (Minimal), OpenAI; 2回の試行に基づくモデル:GPT-Realtime-2.1 High, OpenAI

Speech Agent Arena

Arena 選好 Elo

Preference Elo measures human participants' overall preference after they complete the same scenario with two unidentified Speech to Speech models, such as booking a dental appointment · Only includes models with tool calling · Higher is better

タスク成功率

Proportion of eligible conversations with the correct final task-completing tool call or calls · Higher is better

会話ダイナミクス

会話ダイナミクス(Full Duplex Bench のサブセット)

Weighted average of pause handling, turn-taking, interruption handling, and backchannel handling from Full Duplex Bench v1 and v1.5 · Higher is better

会話ダイナミクス(Full Duplex Bench のサブセット)— カテゴリ別内訳

カテゴリ別スコア(会話ダイナミクススコア順) · Full Duplex Bench v1 および v1.5 のサブセットに基づく

APIベンチマーク

入力音声1時間あたりのコスト(Big Bench Audio のサブセット)

Cost to complete fixed 40-question Big Bench Audio subset based on length of input audio, normalized to hourly basis · Lower is better

音声推論(Big Bench Audio)vs. 入力音声1時間あたりのコスト

Big Bench Audio スコア vs. 入力音声1時間あたりのコスト(固定の40問サブセットで測定) · 音声推論が高く、コストが低いほど良い
Most attractive quadrant

最初の音声までの時間

Average time to first audio (seconds) on Big Bench Audio dataset · Lower is better

音声推論(Big Bench Audio)vs. 速度

Big Bench Audio スコア vs. 最初の音声までの時間 · 音声推論が高く、遅延が低いほど良い
Most attractive quadrant

主要指標の概要と詳細情報

Alibaba Cloud のロゴAlibaba Cloud
Qwen Audio 3.0 Realtime Plus
69.5
99%
98.4%
54.6%
775
77.8%
1.54
4.42
0.03
0.18
Alibaba Cloud のロゴAlibaba Cloud
Qwen3.5 Omni Plus Realtime
-
99%
-
-
859
62.0%
2.64
0.00
0.16
0.82
Alibaba Cloud のロゴAlibaba Cloud
Qwen Audio 3.0 Realtime Flash
67.7
96%
96.9%
35.9%
850
81.7%
1.55
4.77
-
-
Alibaba Cloud のロゴAlibaba Cloud
Qwen3.5 Omni Flash Realtime
-
59%
-
-
841
29.1%
0.79
0.00
0.16
0.82
Alibaba Cloud のロゴAlibaba Cloud
Qwen3 Omni Flash
-
59%
72.7%
-
-
-
4.82
1.77
0.61
1.36
Alibaba Cloud のロゴAlibaba Cloud
Qwen3 Omni Realtime
-
57%
-
-
-
-
0.88
2.26
0.65
0.99
Amazon Bedrock のロゴAmazon Bedrock
Nova 2.0 Sonic (Mar 2026)
-
88%
-
-
974
57.1%
1.14
4.89
0.27
1.08
Boson AI のロゴBoson AI
Higgs Realtime
-
69%
92.8%
18.6%
-
-
1.47
-
-
-
Cofe AI のロゴCofe AI
FLM-Audio
-
16%
62.0%
-
-
-
-
-
-
-
Deepslate のロゴDeepslate
Deepslate Opal
-
85%
85.7%
17.5%
-
-
0.44
6.48
-
-
Google のロゴGoogle
Gemini 3.8 Live Extended Thinking (High)
82.6
98%
91.9%
68.6%
990
89.1%
1.35
3.50
-
-
Google のロゴGoogle
Gemini 3.1 Flash Live High
72.0
97%
74.3%
37.7%
1063
71.8%
2.99
1.75
0.35
1.38
Google のロゴGoogle
Gemini 3.8 Live
76.0
92%
96.1%
30.1%
1083
93.2%
1.18
0.84
-
-
Google のロゴGoogle
Gemini 2.5 Flash Native Audio Dialog Thinking
-
91%
-
-
-
-
3.87
-
0.35
1.38
Google のロゴGoogle
Gemini 3.1 Flash Live Minimal
64.2
71%
72.3%
26.2%
1096
74.6%
0.96
1.50
0.35
1.38
Google のロゴGoogle
Gemini 2.5 Flash Native Audio Dialog
-
69%
-
-
-
-
0.63
1.42
0.35
1.38
Google のロゴGoogle
Gemini 2.5 Flash Native Audio Preview (Dec 2025)
-
-
44.0%
22.8%
-
-
-
-
-
-
Google のロゴGoogle
Gemini 2.5 Flash Native Audio Preview (Sep 2025)
-
-
30.3%
-
-
-
-
-
-
-
Krafton のロゴKrafton
Raon SpeechChat
-
58%
86.2%
-
-
-
0.04
-
-
-
Kyutai のロゴKyutai
Moshi
-
4%
61.0%
-
-
-
-
-
-
-
NVIDIA のロゴNVIDIA
Nemotron Voicechat
-
27%
52.9%
-
-
-
-
-
-
-
NVIDIA のロゴNVIDIA
PersonaPlex
-
19%
91.0%
-
-
-
-
-
-
-
OpenAI のロゴOpenAI
GPT-Realtime-2 (High)
73.6
97%
95.3%
39.8%
932
89.8%
1.14
4.14
1.15
4.61
OpenAI のロゴOpenAI
GPT-Realtime-2.1 High
75.2
96%
95.7%
45.7%
928
91.5%
1.21
10.75
-
-
OpenAI のロゴOpenAI
GPT-Realtime-2 (Medium)
-
93%
95.2%
37.4%
-
-
1.22
3.97
1.15
4.61
OpenAI のロゴOpenAI
GPT-Live-1 (Astra, medium)
83.2
90%
94.9%
74.5%
1048
87.4%
1.34
5.83
-
-
OpenAI のロゴOpenAI
GPT-Live-1 (Sol, low)
80.1
89%
97.3%
59.3%
1053
90.9%
1.24
4.47
-
-
OpenAI のロゴOpenAI
GPT-Realtime-2.1 Minimal
70.7
87%
92.7%
38.0%
937
89.4%
0.97
11.31
-
-
OpenAI のロゴOpenAI
GPT Realtime (Aug '25)
69.1
83%
93.9%
30.4%
974
89.4%
0.98
11.08
1.15
4.61
OpenAI のロゴOpenAI
GPT-Realtime-1.5
70.3
81%
95.7%
38.8%
1000
85.1%
0.81
11.44
1.15
4.61
OpenAI のロゴOpenAI
GPT-Realtime-2.1 Mini High
-
75%
91.7%
29.4%
-
-
4.28
3.45
-
-
OpenAI のロゴOpenAI
GPT-Realtime-2 (Minimal)
63.3
72%
96.1%
30.8%
916
84.7%
1.12
3.07
1.15
4.61
OpenAI のロゴOpenAI
GPT-4o mini Realtime (Dec 2024)
-
69%
-
-
-
-
1.27
5.75
0.36
1.44
OpenAI のロゴOpenAI
GPT Realtime Mini (Oct '25)
56.1
64%
95.7%
15.1%
917
79.6%
0.81
3.04
0.36
1.44
OpenAI のロゴOpenAI
GPT-Realtime-2.1 Mini Minimal
54.4
63%
91.8%
22.5%
840
76.7%
0.85
4.60
-
-
OpenAI のロゴOpenAI
GPT-4o audio chatcompletions
-
54%
-
-
-
-
3.38
0.00
3.60
14.40
OpenAI のロゴOpenAI
GPT-4o Realtime (Dec 2024)
-
-
89.8%
27.9%
-
-
1.49
2.04
1.44
5.76
SpaceXAI のロゴSpaceXAI
Grok Voice Think Fast 2.0 High
81.3
97%
95.1%
56.5%
1011
94.6%
0.70
4.80
-
-
SpaceXAI のロゴSpaceXAI
Grok Voice Think Fast 1.0
73.7
97%
77.8%
52.1%
908
80.7%
1.25
3.00
-
-
SpaceXAI のロゴSpaceXAI
Grok Voice Fast 1.0
-
93%
71.6%
27.4%
-
-
0.78
3.00
3.00
3.00
StepFun のロゴStepFun
StepAudio 3 Realtime
-
100%
98.9%
-
-
-
8.83
-
-
-
StepFun のロゴStepFun
Step-Audio R1.1 (Realtime)
-
98%
-
-
-
-
1.53
0.00
0.06
1.69
VITA のロゴVITA
Freeze-Omni
-
33%
58.7%
-
-
-
-
-
-
-

よくある質問