Vendor page

Speech to Text (ASR) Providers Leaderboard & Comparison

Artificial Analysis, Artificial Analysis, 2026

Read the original

Read on October 19, 2026

What it is: The public speech-to-text leaderboard of Artificial Analysis, an independent benchmarking company. Its summary table gives each transcription provider's word error rate on the AA-WER index (non-streaming), with speed factor and price per 1,000 minutes. The index is a weighted average over three test datasets.
How much weight it carries: A vendor-run benchmark that publishes its datasets, weights and method, but one of its three datasets (AA-AgentTalk) is proprietary and cannot be checked by outsiders. The page is updated over time, so figures change: the 2026 values here are the ones shown on the day of reading, and the article of February 2026 gave different figures for some systems.

Facts taken from this source

On Artificial Analysis's speech-to-text leaderboard, Amazon Transcribe Amazon Bedrock had an AA-WER of 4.1 percent, its word error rate on the index's three test datasets.
“Amazon Transcribe Amazon Bedrock 4.1% 31.3 6.00 Details” (Summary of Key Metrics table)
On Artificial Analysis's speech-to-text leaderboard, Canary Qwen 2.5B, NVIDIA Replicate had an AA-WER of 4.3 percent, its word error rate on the index's three test datasets.
“Canary Qwen 2.5B, NVIDIA Replicate 4.3% 8.8 0.74 Details” (Summary of Key Metrics table)
On Artificial Analysis's speech-to-text leaderboard, Chirp 3, Google Google had an AA-WER of 4.3 percent, its word error rate on the index's three test datasets.
“Chirp 3, Google Google 4.3% 29.5 16.00 Details” (Summary of Key Metrics table)
On Artificial Analysis's speech-to-text leaderboard, Chirp Google had an AA-WER of 31.2 percent, its word error rate on the index's three test datasets.
“Chirp Google 31.2% 14.0 16.00 Details” (Summary of Key Metrics table)
On Artificial Analysis's speech-to-text leaderboard, Fun-Realtime-ASR-preview Alibaba Cloud had an AA-WER of 1.7 percent, its word error rate on the index's three test datasets.
“Fun-Realtime-ASR-preview Alibaba Cloud 1.7% 20.4 0.00 Details” (Summary of Key Metrics table)
On Artificial Analysis's speech-to-text leaderboard, Gemini 2.5 Flash Google had an AA-WER of 5.1 percent, its word error rate on the index's three test datasets.
“Gemini 2.5 Flash Google 5.1% 71.5 6.66 Details” (Summary of Key Metrics table)
On Artificial Analysis's speech-to-text leaderboard, Gemini 2.5 Flash Lite Google had an AA-WER of 5.2 percent, its word error rate on the index's three test datasets.
“Gemini 2.5 Flash Lite Google 5.2% 76.7 6.56 Details” (Summary of Key Metrics table)
On Artificial Analysis's speech-to-text leaderboard, Gemini 2.5 Pro Google had an AA-WER of 2.9 percent, its word error rate on the index's three test datasets.
“Gemini 2.5 Pro Google 2.9% 10.6 11.39 Details” (Summary of Key Metrics table)
On Artificial Analysis's speech-to-text leaderboard, Gemini 3.1 Flash-Lite Preview (Minimal) Google had an AA-WER of 3.4 percent, its word error rate on the index's three test datasets.
“Gemini 3.1 Flash-Lite Preview (Minimal) Google 3.4% 88.5 5.83 Details” (Summary of Key Metrics table)
On Artificial Analysis's speech-to-text leaderboard, Gemini 3.1 Pro Preview (High) Google had an AA-WER of 2.8 percent, its word error rate on the index's three test datasets.
“Gemini 3.1 Pro Preview (High) Google 2.8% 7.8 18.15 Details” (Summary of Key Metrics table)
On Artificial Analysis's speech-to-text leaderboard, Gemini 3.1 Pro Preview (Low) Google had an AA-WER of 3.6 percent, its word error rate on the index's three test datasets.
“Gemini 3.1 Pro Preview (Low) Google 3.6% 7.7 7.72 Details” (Summary of Key Metrics table)
On Artificial Analysis's speech-to-text leaderboard, Gemini 3.5 Transcribe Google had an AA-WER of 2.6 percent, its word error rate on the index's three test datasets.
“Gemini 3.5 Transcribe Google 2.6% 91.7 5.00 Details” (Summary of Key Metrics table)
On Artificial Analysis's speech-to-text leaderboard, Gemini 3 Flash (High) Google had an AA-WER of 2.9 percent, its word error rate on the index's three test datasets.
“Gemini 3 Flash (High) Google 2.9% 20.0 13.70 Details” (Summary of Key Metrics table)
On Artificial Analysis's speech-to-text leaderboard, GPT-4o Mini Transcribe OpenAI had an AA-WER of 4.5 percent, its word error rate on the index's three test datasets.
“GPT-4o Mini Transcribe OpenAI 4.5% 52.3 3.00 Details” (Summary of Key Metrics table)
On Artificial Analysis's speech-to-text leaderboard, GPT-4o Transcribe OpenAI had an AA-WER of 4.0 percent, its word error rate on the index's three test datasets.
“GPT-4o Transcribe OpenAI 4.0% 37.9 6.00 Details” (Summary of Key Metrics table)
On Artificial Analysis's speech-to-text leaderboard, GPT Transcribe, OpenAI OpenAI had an AA-WER of 3.3 percent, its word error rate on the index's three test datasets.
“GPT Transcribe, OpenAI OpenAI 3.3% 56.3 4.50 Details” (Summary of Key Metrics table)
On Artificial Analysis's speech-to-text leaderboard, Gradium Speech-to-Text Gradium had an AA-WER of 6.8 percent, its word error rate on the index's three test datasets.
“Gradium Speech-to-Text Gradium 6.8% 2.3 13.00 Details” (Summary of Key Metrics table)
On Artificial Analysis's speech-to-text leaderboard, Grok Voice Transcribe 2.0 SpaceXAI had an AA-WER of 2.3 percent, its word error rate on the index's three test datasets.
“Grok Voice Transcribe 2.0 SpaceXAI 2.3% 158.5 1.67 Details” (Summary of Key Metrics table)
On Artificial Analysis's speech-to-text leaderboard, Incredibly Fast Whisper Replicate large-v3 had an AA-WER of 5.7 percent, its word error rate on the index's three test datasets.
“Incredibly Fast Whisper Replicate large-v3 5.7% 54.1 1.49 Details” (Summary of Key Metrics table)
On Artificial Analysis's speech-to-text leaderboard, Inworld STT 1 Inworld had an AA-WER of 3.9 percent, its word error rate on the index's three test datasets.
“Inworld STT 1 Inworld 3.9% 147.1 2.50 Details” (Summary of Key Metrics table)
On Artificial Analysis's speech-to-text leaderboard, MAI-Transcribe-1.5 Microsoft AI had an AA-WER of 2.4 percent, its word error rate on the index's three test datasets.
“MAI-Transcribe-1.5 Microsoft AI 2.4% 80.5 6.00 Details” (Summary of Key Metrics table)
On Artificial Analysis's speech-to-text leaderboard, MAI-Transcribe-2 Microsoft AI had an AA-WER of 2.0 percent, its word error rate on the index's three test datasets.
“MAI-Transcribe-2 Microsoft AI 2.0% 317.0 1.67 Details” (Summary of Key Metrics table)
On Artificial Analysis's speech-to-text leaderboard, Modulate STT Batch English VFast Modulate had an AA-WER of 4.2 percent, its word error rate on the index's three test datasets.
“Modulate STT Batch English VFast Modulate 4.2% 47.8 0.42 Details” (Summary of Key Metrics table)
On Artificial Analysis's speech-to-text leaderboard, Nova 2 Pro Amazon Bedrock had an AA-WER of 4.9 percent, its word error rate on the index's three test datasets.
“Nova 2 Pro Amazon Bedrock 4.9% 23.8 3.10 Details” (Summary of Key Metrics table)
On Artificial Analysis's speech-to-text leaderboard, Nova-2, Telnyx Telnyx had an AA-WER of 5.1 percent, its word error rate on the index's three test datasets.
“Nova-2, Telnyx Telnyx 5.1% 459.6 7.40 Details” (Summary of Key Metrics table)
On Artificial Analysis's speech-to-text leaderboard, Nova-3 Deepgram had an AA-WER of 5.2 percent, its word error rate on the index's three test datasets.
“Nova-3 Deepgram 5.2% 625.7 4.30 Details” (Summary of Key Metrics table)
On Artificial Analysis's speech-to-text leaderboard, Nova-3, Telnyx Telnyx had an AA-WER of 4.8 percent, its word error rate on the index's three test datasets.
“Nova-3, Telnyx Telnyx 4.8% 336.1 7.40 Details” (Summary of Key Metrics table)
On Artificial Analysis's speech-to-text leaderboard, Parakeet RNNT 1.1B Replicate had an AA-WER of 5.4 percent, its word error rate on the index's three test datasets.
“Parakeet RNNT 1.1B Replicate 5.4% 6.4 1.91 Details” (Summary of Key Metrics table)
On Artificial Analysis's speech-to-text leaderboard, Parakeet TDT 0.6B V2, NVIDIA NVIDIA had an AA-WER of 6.4 percent, its word error rate on the index's three test datasets.
“Parakeet TDT 0.6B V2, NVIDIA NVIDIA 6.4% 95.2 0.00 Details” (Summary of Key Metrics table)
On Artificial Analysis's speech-to-text leaderboard, Parakeet TDT 0.6B V3, Togetherai Together AI had an AA-WER of 4.5 percent, its word error rate on the index's three test datasets.
“Parakeet TDT 0.6B V3, Togetherai Together AI 4.5% 527.4 1.50 Details” (Summary of Key Metrics table)
On Artificial Analysis's speech-to-text leaderboard, Qwen3.5 Omni Flash Alibaba Cloud had an AA-WER of 13.5 percent, its word error rate on the index's three test datasets.
“Qwen3.5 Omni Flash Alibaba Cloud 13.5% 74.4 0.00 Details” (Summary of Key Metrics table)
On Artificial Analysis's speech-to-text leaderboard, Qwen3.5 Omni Plus Alibaba Cloud had an AA-WER of 3.5 percent, its word error rate on the index's three test datasets.
“Qwen3.5 Omni Plus Alibaba Cloud 3.5% 166.7 0.00 Details” (Summary of Key Metrics table)
On Artificial Analysis's speech-to-text leaderboard, Resonant-1 Reson8 had an AA-WER of 3.4 percent, its word error rate on the index's three test datasets.
“Resonant-1 Reson8 3.4% 327.4 3.60 Details” (Summary of Key Metrics table)
On Artificial Analysis's speech-to-text leaderboard, Rev AI Rev AI had an AA-WER of 5.9 percent, its word error rate on the index's three test datasets.
“Rev AI Rev AI 5.9% 13.1 3.33 Details” (Summary of Key Metrics table)
On Artificial Analysis's speech-to-text leaderboard, Scribe v2 ElevenLabs had an AA-WER of 2.2 percent, its word error rate on the index's three test datasets.
“Scribe v2 ElevenLabs 2.2% 63.0 3.67 Details” (Summary of Key Metrics table)
On Artificial Analysis's speech-to-text leaderboard, Smallest AI Pulse Pro Smallest.ai had an AA-WER of 3.4 percent, its word error rate on the index's three test datasets.
“Smallest AI Pulse Pro Smallest.ai 3.4% 291.6 4.00 Details” (Summary of Key Metrics table)
On Artificial Analysis's speech-to-text leaderboard, Smallest AI Pulse Smallest.ai had an AA-WER of 3.3 percent, its word error rate on the index's three test datasets.
“Smallest AI Pulse Smallest.ai 3.3% 259.3 5.00 Details” (Summary of Key Metrics table)
On Artificial Analysis's speech-to-text leaderboard, Solaria-1, Gladia Gladia had an AA-WER of 4.1 percent, its word error rate on the index's three test datasets.
“Solaria-1, Gladia Gladia 4.1% 90.3 10.17 Details” (Summary of Key Metrics table)
On Artificial Analysis's speech-to-text leaderboard, Solaria-3, Gladia Gladia had an AA-WER of 3.2 percent, its word error rate on the index's three test datasets.
“Solaria-3, Gladia Gladia 3.2% 91.8 10.16 Details” (Summary of Key Metrics table)
On Artificial Analysis's speech-to-text leaderboard, Soniox v5 Async Soniox had an AA-WER of 3.8 percent, its word error rate on the index's three test datasets.
“Soniox v5 Async Soniox 3.8% 43.5 1.66 Details” (Summary of Key Metrics table)
On Artificial Analysis's speech-to-text leaderboard, Speechmatics Enhanced Speechmatics had an AA-WER of 4.0 percent, its word error rate on the index's three test datasets.
“Speechmatics Enhanced Speechmatics 4.0% 69.8 12.50 Details” (Summary of Key Metrics table)
On Artificial Analysis's speech-to-text leaderboard, Speechmatics Melia Speechmatics had an AA-WER of 4.9 percent, its word error rate on the index's three test datasets.
“Speechmatics Melia Speechmatics 4.9% 167.9 4.00 Details” (Summary of Key Metrics table)
On Artificial Analysis's speech-to-text leaderboard, Speechmatics Standard Speechmatics had an AA-WER of 5.1 percent, its word error rate on the index's three test datasets.
“Speechmatics Standard Speechmatics 5.1% 100.4 7.50 Details” (Summary of Key Metrics table)
On Artificial Analysis's speech-to-text leaderboard, StepAudio 2.5 ASR, StepFun StepFun had an AA-WER of 4.7 percent, its word error rate on the index's three test datasets.
“StepAudio 2.5 ASR, StepFun StepFun 4.7% 122.4 0.37 Details” (Summary of Key Metrics table)
On Artificial Analysis's speech-to-text leaderboard, StepAudio 3 ASR, StepFun StepFun had an AA-WER of 2.5 percent, its word error rate on the index's three test datasets.
“StepAudio 3 ASR, StepFun StepFun 2.5% 79.4 6.67 Details” (Summary of Key Metrics table)
On Artificial Analysis's speech-to-text leaderboard, transcribe-03-2026 Cohere had an AA-WER of 4.6 percent, its word error rate on the index's three test datasets.
“transcribe-03-2026 Cohere 4.6% 117.2 0.00 Details” (Summary of Key Metrics table)
On Artificial Analysis's speech-to-text leaderboard, Universal-3.5 Pro AssemblyAI had an AA-WER of 3.0 percent, its word error rate on the index's three test datasets.
“Universal-3.5 Pro AssemblyAI 3.0% 105.5 3.50 Details” (Summary of Key Metrics table)
On Artificial Analysis's speech-to-text leaderboard, Universal-3 Pro AssemblyAI had an AA-WER of 3.1 percent, its word error rate on the index's three test datasets.
“Universal-3 Pro AssemblyAI 3.1% 109.9 3.50 Details” (Summary of Key Metrics table)
On Artificial Analysis's speech-to-text leaderboard, Universal, AssemblyAI AssemblyAI had an AA-WER of 3.8 percent, its word error rate on the index's three test datasets.
“Universal, AssemblyAI AssemblyAI 3.8% 119.1 2.50 Details” (Summary of Key Metrics table)
On Artificial Analysis's speech-to-text leaderboard, Voxtral Mini DeepInfra had an AA-WER of 3.8 percent, its word error rate on the index's three test datasets.
“Voxtral Mini DeepInfra 3.8% 51.3 1.00 Details” (Summary of Key Metrics table)
On Artificial Analysis's speech-to-text leaderboard, Voxtral Mini Transcribe 2 Mistral had an AA-WER of 3.6 percent, its word error rate on the index's three test datasets.
“Voxtral Mini Transcribe 2 Mistral 3.6% 86.1 3.00 Details” (Summary of Key Metrics table)
On Artificial Analysis's speech-to-text leaderboard, Voxtral Small Mistral had an AA-WER of 2.8 percent, its word error rate on the index's three test datasets.
“Voxtral Small Mistral 2.8% 61.5 4.00 Details” (Summary of Key Metrics table)
On Artificial Analysis's speech-to-text leaderboard, Whisper Large v2 OpenAI large-v2 had an AA-WER of 4.1 percent, its word error rate on the index's three test datasets.
“Whisper Large v2 OpenAI large-v2 4.1% 28.5 6.00 Details” (Summary of Key Metrics table)
On Artificial Analysis's speech-to-text leaderboard, Whisper Large v3 fal.ai large-v3 had an AA-WER of 4.1 percent, its word error rate on the index's three test datasets.
“Whisper Large v3 fal.ai large-v3 4.1% 104.5 1.15 Details” (Summary of Key Metrics table)
On Artificial Analysis's speech-to-text leaderboard, Whisper Large v3 Replicate large-v3 had an AA-WER of 10.1 percent, its word error rate on the index's three test datasets.
“Whisper Large v3 Replicate large-v3 10.1% 2.7 4.23 Details” (Summary of Key Metrics table)
On Artificial Analysis's speech-to-text leaderboard, Whisper Large v3 Together AI large-v3 had an AA-WER of 4.5 percent, its word error rate on the index's three test datasets.
“Whisper Large v3 Together AI large-v3 4.5% 290.2 1.50 Details” (Summary of Key Metrics table)
On Artificial Analysis's speech-to-text leaderboard, Whisper Large v3 Turbo Groq v3 Turbo had an AA-WER of 4.6 percent, its word error rate on the index's three test datasets.
“Whisper Large v3 Turbo Groq v3 Turbo 4.6% 195.0 0.67 Details” (Summary of Key Metrics table)
On Artificial Analysis's speech-to-text leaderboard, Whisper Large v3 Turbo, Telnyx Telnyx had an AA-WER of 5.5 percent, its word error rate on the index's three test datasets.
“Whisper Large v3 Turbo, Telnyx Telnyx 5.5% 37.8 15.00 Details” (Summary of Key Metrics table)
On Artificial Analysis's speech-to-text leaderboard, Wizper Large v3 fal.ai large-v3 had an AA-WER of 4.7 percent, its word error rate on the index's three test datasets.
“Wizper Large v3 fal.ai large-v3 4.7% 184.6 0.50 Details” (Summary of Key Metrics table)
On Artificial Analysis's speech-to-text leaderboard, Zoom Scribe V2 Pro Zoom had an AA-WER of 2.8 percent, its word error rate on the index's three test datasets.
“Zoom Scribe V2 Pro Zoom 2.8% 121.3 3.33 Details” (Summary of Key Metrics table)

Pages that rest on this source

Speech Recognition Accuracy: Word Error RatesDataset