음성 대 음성 AI 모델 및 제공업체 리더보드

음성 대 음성 모델과 API 제공업체를 분석하고 비교합니다. Artificial Analysis는 추론 품질, 대화 역학, 생성 시간, 가격 등 다양한 특성을 기준으로 음성 대 음성 모델과 호스팅 제공업체를 분석했습니다.

자세한 내용은 방법론 페이지를 참고하세요.

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

음성 추론

음성 추론(Big Bench Audio)

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

에이전트 성능

에이전트 성능(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

참고: 1회 시행 기준 모델: GPT-Live-1 (Astra, medium), OpenAI, GPT-Live-1 (Sol, low), OpenAI, GPT-Realtime-2 (Minimal), OpenAI; 2회 시행 기준 모델: GPT-Realtime-2.1 High, OpenAI

도메인별 에이전트 성능(𝜏-Voice)

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

참고: 1회 시행 기준 모델: GPT-Live-1 (Astra, medium), OpenAI, GPT-Live-1 (Sol, low), OpenAI, GPT-Realtime-2 (Minimal), OpenAI; 2회 시행 기준 모델: GPT-Realtime-2.1 High, OpenAI

Speech Agent Arena

Arena 선호도 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

작업 성공률

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

대화 역학

대화 역학(Full Duplex Bench 하위 집합)

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

대화 역학(Full Duplex Bench 하위 집합) - 카테고리별 세부 내역

대화 역학 점수 순으로 정렬한 카테고리별 점수 · Full Duplex Bench v1 및 v1.5 하위 집합 기준

API 벤치마크

입력 오디오 시간당 비용(Big Bench Audio 하위 집합)

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

음성 추론(Big Bench Audio) vs. 입력 오디오 시간당 비용

Big Bench Audio 점수 vs. 입력 오디오 시간당 비용(고정된 40문항 하위 집합에서 측정) · 음성 추론이 높고 비용이 낮을수록 좋음
Most attractive quadrant

최초 오디오 생성 시간

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

음성 추론(Big Bench Audio) vs. 속도

Big Bench Audio 점수 vs. 최초 오디오 생성 시간 · 음성 추론이 높고 지연 시간이 낮을수록 좋음
Most attractive quadrant

주요 지표 요약 및 추가 정보

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

자주 묻는 질문