The ranker
Every model. Same audio. Same metrics. Re-run on every model or pricing change.
10 free transcriptions, up to 2 hours on signup — no card.
40
Models Ranked
1,299
Total Benchmarks
11
Languages Tested
Nov 5, 2026
Last Updated
01 · Leader
MAI-Transcribe-2 (preview)
Microsoft Azure
89.9
Score
WER
7.77%
Latency
946ms
Cost
$0.0017/min
02 · Runner-up
Grok Voice Transcribe 2.0
xAI
88.1
Score
WER
9.95%
Latency
1.2s
Cost
$0.0017/min
03 · Third
Whisper Large V3 Turbo
Groq
83.8
Score
WER
14.47%
Latency
2.5s
Cost
$0.0007/min
| # | Model | Overall Score | WER | CER | Latency | Cost | Benchmarks |
|---|---|---|---|---|---|---|---|
01 | MAI-Transcribe-2 (preview) Microsoft Azure | 89.9 | 7.77% | 6.06% | 946ms | $0.0017/min | 35 |
02 | Grok Voice Transcribe 2.0 xAI | 88.1 | 9.95% | 7.49% | 1.2s | $0.0017/min | 35 |
03 | Whisper Large V3 Turbo Groq | 83.8 | 14.47% | 10.63% | 2.5s | $0.0007/min | 33 |
04 | GPT-4o Mini Transcribe OpenAI | 81.1 | 13.58% | 9.32% | 2.0s | $0.003/min | 35 |
05 | Whisper Large V3 Groq | 80.1 | 14.89% | 11.36% | 2.9s | $0.0019/min | 33 |
06 | GPT Transcribe OpenAI | 79.2 | 9.58% | 7.11% | 2.3s | $0.0045/min | 35 |
07 | Scribe v2 ElevenLabs | 78.2 | 9.15% | 7.19% | 3.1s | $0.004/min | 35 |
08 | Whisper Large V3 Turbo Cloudflare | 76.7 | 12.90% | 9.21% | 5.3s | $0.0005/min | 33 |
09 | Nova-2 Voicemail Deepgram | 76.7 | 17.51% | 12.36% | 954ms | $0.0058/min | 25 |
10 | Nova-2 Deepgram | 76.2 | 16.54% | 11.93% | 1.3s | $0.0058/min | 34 |
11 | Nova-2 Phone Call Deepgram | 75.7 | 16.38% | 12.42% | 1.5s | $0.0058/min | 25 |
12 | Voxtral Mini Transcribe Mistral (Voxtral) | 75.4 | 12.01% | 7.68% | 4.2s | $0.003/min | 30 |
13 | Azure Speech Microsoft Azure | 74.4 | 15.34% | 10.67% | 2.0s | $0.006/min | 35 |
14 | Nova-3 Deepgram | 74.3 | 15.86% | 10.95% | 787ms | $0.0077/min | 35 |
15 | GPT-4o Transcribe OpenAI | 74.0 | 14.76% | 11.60% | 2.2s | $0.006/min | 35 |
16 | Soniox STT (Async) Soniox | 73.7 | 8.73% | 6.18% | 6.2s | $0.0017/min | 34 |
17 | Nova-2 Meeting Deepgram | 71.3 | 16.74% | 11.73% | 2.9s | $0.0058/min | 25 |
18 | Nova-2 Finance Deepgram | 69.6 | 18.92% | 13.09% | 3.1s | $0.0058/min | 25 |
19 | Gemini 3.5 Transcribe Google Gemini | 69.4 | 10.43% | 6.94% | 5.1s | $0.005/min | 35 |
20 | Whisper Large V3 OpenAI | 69.3 | 17.67% | 13.06% | 3.3s | $0.006/min | 35 |
21 | Qwen3 ASR Flash Alibaba (Qwen) | 69.1 | 12.27% | 7.39% | 6.9s | $0.0021/min | 35 |
22 | Whisper 1 (API) OpenAI | 69.0 | 17.63% | 13.03% | 3.4s | $0.006/min | 35 |
23 | Nova-2 Conversational AI Deepgram | 67.6 | 18.23% | 13.27% | 3.9s | $0.0058/min | 25 |
24 | Universal-3.5 Pro AssemblyAI | 67.4 | 17.44% | 10.37% | 5.6s | $0.0035/min | 35 |
25 | Base Deepgram | 65.2 | 25.04% | 18.09% | 772ms | $0.0145/min | 34 |
26 | Enhanced Deepgram | 63.8 | 24.86% | 14.27% | 1.3s | $0.0165/min | 32 |
27 | Muse Voice Transcribe Meta | 61.8 | 14.22% | 10.91% | 8.4s | $0.003/min | 34 |
28 | Speechmatics Standard Speechmatics | 60.9 | 14.69% | 9.80% | 7.3s | $0.005/min | 35 |
29 | Rev AI Reverb Rev AI | 54.9 | 18.18% | 13.46% | 13.7s | $0.003/min | 35 |
30 | Gladia Solaria-3 Gladia | 52.2 | 9.80% | 5.76% | 7.6s | $0.0101/min | 20 |
31 | Amazon Transcribe Amazon Web Services | 51.5 | 13.02% | 9.17% | 12.4s | $0.006/min | 35 |
32 | GPT-4o Transcribe Diarize OpenAI | 50.7 | 14.61% | 9.97% | 15.2s | $0.006/min | 33 |
33 | Speechmatics Enhanced Speechmatics | 47.3 | 12.30% | 7.96% | 12.1s | $0.0083/min | 35 |
34 | Chirp 3 Google Cloud | 45.0 | 9.95% | 5.81% | 11.0s | $0.0107/min | 34 |
35 | Amazon Transcribe Medical Amazon Web Services | 42.5 | 14.92% | 9.67% | 11.4s | $0.075/min | 25 |
36 | Google Telephony Google Cloud | 39.9 | 20.24% | 12.59% | 21.0s | $0.016/min | 32 |
37 | Google Latest (Long) Google Cloud | 38.7 | 22.55% | 13.81% | 13.2s | $0.0107/min | 34 |
38 | Google Default Google Cloud | 32.7 | 34.65% | 25.66% | 13.4s | $0.016/min | 35 |
39 | Google Command & Search Google Cloud | 32.4 | 35.22% | 26.46% | 12.7s | $0.016/min | 35 |
40 | Google Latest (Short) Google Cloud | 21.1 | 57.82% | 52.44% | 11.3s | $0.016/min | 34 |
01
MAI-Transcribe-2 (preview)
89.9
7.77% WER
946ms
$0.0017/min
02
Grok Voice Transcribe 2.0
88.1
9.95% WER
1.2s
$0.0017/min
03
Whisper Large V3 Turbo
83.8
14.47% WER
2.5s
$0.0007/min
04
GPT-4o Mini Transcribe
81.1
13.58% WER
2.0s
$0.003/min
05
Whisper Large V3
80.1
14.89% WER
2.9s
$0.0019/min
06
GPT Transcribe
79.2
9.58% WER
2.3s
$0.0045/min
07
Scribe v2
78.2
9.15% WER
3.1s
$0.004/min
08
Whisper Large V3 Turbo
76.7
12.90% WER
5.3s
$0.0005/min
09
Nova-2 Voicemail
76.7
17.51% WER
954ms
$0.0058/min
10
Nova-2
76.2
16.54% WER
1.3s
$0.0058/min
11
Nova-2 Phone Call
75.7
16.38% WER
1.5s
$0.0058/min
12
Voxtral Mini Transcribe
75.4
12.01% WER
4.2s
$0.003/min
13
Azure Speech
74.4
15.34% WER
2.0s
$0.006/min
14
Nova-3
74.3
15.86% WER
787ms
$0.0077/min
15
GPT-4o Transcribe
74.0
14.76% WER
2.2s
$0.006/min
16
Soniox STT (Async)
73.7
8.73% WER
6.2s
$0.0017/min
17
Nova-2 Meeting
71.3
16.74% WER
2.9s
$0.0058/min
18
Nova-2 Finance
69.6
18.92% WER
3.1s
$0.0058/min
19
Gemini 3.5 Transcribe
69.4
10.43% WER
5.1s
$0.005/min
20
Whisper Large V3
69.3
17.67% WER
3.3s
$0.006/min
21
Qwen3 ASR Flash
69.1
12.27% WER
6.9s
$0.0021/min
22
Whisper 1 (API)
69.0
17.63% WER
3.4s
$0.006/min
23
Nova-2 Conversational AI
67.6
18.23% WER
3.9s
$0.0058/min
24
Universal-3.5 Pro
67.4
17.44% WER
5.6s
$0.0035/min
25
Base
65.2
25.04% WER
772ms
$0.0145/min
26
Enhanced
63.8
24.86% WER
1.3s
$0.0165/min
27
Muse Voice Transcribe
61.8
14.22% WER
8.4s
$0.003/min
28
Speechmatics Standard
60.9
14.69% WER
7.3s
$0.005/min
29
Rev AI Reverb
54.9
18.18% WER
13.7s
$0.003/min
30
Gladia Solaria-3
52.2
9.80% WER
7.6s
$0.0101/min
31
Amazon Transcribe
51.5
13.02% WER
12.4s
$0.006/min
32
GPT-4o Transcribe Diarize
50.7
14.61% WER
15.2s
$0.006/min
33
Speechmatics Enhanced
47.3
12.30% WER
12.1s
$0.0083/min
34
Chirp 3
45.0
9.95% WER
11.0s
$0.0107/min
35
Amazon Transcribe Medical
42.5
14.92% WER
11.4s
$0.075/min
36
Google Telephony
39.9
20.24% WER
21.0s
$0.016/min
37
Google Latest (Long)
38.7
22.55% WER
13.2s
$0.0107/min
38
Google Default
32.7
34.65% WER
13.4s
$0.016/min
39
Google Command & Search
32.4
35.22% WER
12.7s
$0.016/min
40
Google Latest (Short)
21.1
57.82% WER
11.3s
$0.016/min
Code-Switching
Scribe v2
ElevenLabs
n=2 · too few to separate
Conversational
Chirp 3
Google Cloud
n=2 · too few to separate
Finance
Universal-3.5 Pro
AssemblyAI
n=2 · too few to separate
General
Amazon Transcribe Medical
Amazon Web Services
within margin · 1.25%–10.02%
Legal
Azure Speech
Microsoft Azure
n=2 · too few to separate
Medical
Soniox STT (Async)
Soniox
n=2 · too few to separate
Noisy Environment
Grok Voice Transcribe 2.0
xAI
within margin · 1.67%–32.50%
Technical
MAI-Transcribe-2 (preview)
Microsoft Azure
n=2 · too few to separate
50%
Accuracy
WER + CER vs. reference transcripts
30%
Speed
Median end-to-end latency
20%
Cost efficiency
Price per minute of audio
01
Golden set
Curated test audio with verified human reference transcripts across languages, accents, and noise levels. Some public read-speech corpora (LibriSpeech, FLEURS, Common Voice) may appear in model training data, which can flatter the models trained on them; we counter this by also scoring across domain audio — legal, financial, and conversational — that is unlikely to be in standard training sets.
02
Same audio, vendor-recommended settings
Every model runs the identical test set — no cherry-picked clips. Our rule for calls: each model uses its provider's recommended settings — the configuration real users would run — not a lowest-common-denominator call that penalizes a model for its vendor's best practices (e.g. OpenAI's gpt-4o models get server-side VAD chunking).
03
Event-driven & automated
Benchmarks re-run automatically on every model or pricing change, with no human bias. Every provider gets the same test audio.
04
Scoring
Overall score is a weighted composite: 50% accuracy (WER), 30% speed, 20% cost efficiency. Category leaders show a confidence interval; where clips are too few to separate models, we say so rather than crown a winner.
Corpus sources & licenses
The golden set is built from openly-licensed audio. LibriSpeech and FLEURS are CC BY 4.0 and require attribution; the rest is credited for transparency.
LibriSpeech (test-clean) — CC BY 4.0
Mozilla Common Voice — CC0 1.0
Bangor Miami (TalkBank) — GPLv3
Spoken Wikipedia CS Corpus — CC BY-SA 3.0
U.S. government recordings (SCOTUS, NIH/CDC) — Public domain
Original recordings — OpenTranscription
WER
Word Error Rate
Percentage of words incorrectly transcribed (lower is better)
CER
Character Error Rate
Percentage of characters incorrectly transcribed (lower is better)
MER
Match Error Rate
Ratio of errors to total alignment length (lower is better)
WIL
Word Information Lost
Fraction of word information lost in transcription (lower is better)
One endpoint, every provider. Pin the leader or let us auto-route to the best model under your accuracy and latency budget.
