Apple SpeechAnalyzer and Argmax WhisperKit

TL;DR

• In WWDC 2025, Apple introduced SpeechAnalyzer to modernize its on-device speech recognition framework with a new proprietary Apple model.
• In our benchmarks, Apple matches the speed and accuracy of mid-tier OpenAI Whisper models on long-form conversational speech transcription.
• Developers looking for a free offering with this specific mid-tier speed-accuracy trade-off can pick either Apple SpeechAnalyzer or a smaller model in Argmax WhisperKit, depending on their other requirements. We publish a comprehensive feature set comparison below.
• For those with even more demanding requirements, Argmax SDK offers frontier accuracy for speech AND speaker recognition while achieving ~5x higher transcription speed compared to either framework.

‍

Benchmarks

↑: Higher is better

‍

Speed Factor (↑)

Speed factor indicates the number of seconds of input audio processed by the transcription system in one second of wall-clock time, e.g. A speed factor of 60 means that a system can process 1 minute of audio in 1 second.

All results are computed on an M4 Mac mini running macOS 26 Beta Seed 1. Apple results are obtained through this open-source benchmark script and can be easily reproduced. Argmax results are obtained in our Playground app on TestFlight and can be reproduced even more easily.

‍

Error Rate (↓)

This is the Word Error Rate (WER) metric computed on a random 10% subset of the earnings22 dataset, consisting of ~12 hours of English conversations from earnings calls with analysts. The reason for picking this dataset is that Apple mentions long-form conversational speech as the primary improvement with their new SpeechTranscriber model.

‍

Feature Set Comparison

‍

Deployment Considerations

‍

Support Considerations

‍

Commercial Considerations

‍

‍

‍

Keyword Recognition Accuracy

The Error Rate results from above provide high-level insights into the speech-to-text accuracy of Apple and Argmax systems on general vocabulary. However, many real-world use cases disproportionately rely on accuracy for important known keywords instead.

The following benchmark demonstrates the keyword recognition accuracy of Apple SpeechAnalyzer (iOS 26), Apple SFSpeechRecogizer (pre-iOS 26) and Argmax SDK. Keywords are defined as people, company and produt names that occur in the earnings22 dataset. Please see this GitHub repository for details.

‍

‍

Notably, Apple's new SpeechAnalyzer (iOS 26) API lacks the Custom Vocabulary feature that lets developers improve accuracy on known-and-registered keywords while Apple's older SFSpeechRecognizer API (pre-iOS 26) has this feature and surpasses their new API in accuracy.

Argmax with Custom Vocabulary surpasses both by a significant margin and even matches top cloud APIs in keyword recognition accuracy.

‍

Argmax will integrate Apple

We were slightly disappointed to see that Apple’s model still requires a download and does not come pre-installed with iOS or macOS. However, if Apple SpeechAnalyzer is widely adopted, a newly installed app will find that the model was previously downloaded by another app, including Apple’s first-party apps, on the same device, and skip the download! This removes a significant obstacle for on-device deployment: the latency from app install to first inference, which is dominated by model download time.

For this purpose, Argmax will integrate Apple SpeechAnalyzer so that Argmax WhisperKit and Argmax SDK users may also benefit from a pre-downloaded model while their Argmax model is being downloaded for them.

‍

Browse Apple SpeechAnalyzer documentation.

Start with Argmax WhisperKit on GitHub.

Get access to Argmax SDK.

‍

‍