# Hugging Face releases AnySpeechRecognizer and AnyDecisionModel for Apple platforms

Digest AI · Agents & Tools · published 2026-10-29T00:00:00Z

Canonical: https://digestai.news/story/hugging-face-releases-anyspeechrecognizer-and-anydecisionmodel-for-app

## Summary

Hugging Face announced two open-source Swift packages for Apple platforms: **AnySpeechRecognizer** and **AnyDecisionModel**. Both extend Apple’s existing APIs by letting developers choose from multiple speech recognition and decision-making models, including NVIDIA’s Parakeet, Alibaba’s Qwen3-ASR, and Cloudflare’s Clef-Flash, running locally on Core ML or MLX backends.

**AnySpeechRecognizer** integrates speech-to-text and speaker diarization, supporting 25 European languages (Parakeet) or 30 languages plus 22 Chinese dialects (Qwen3-ASR). It mirrors Apple’s SpeechAnalyzer API but allows swapping engines—e.g., replacing Apple’s default with Parakeet v3 or Qwen3-ASR. The package includes a **Sortformer** diarizer for tracking up to four speakers in overlapping speech. Benchmarks on an M4 Max Mac Studio show Parakeet transcribing an hour of audio in **20–40 seconds** (file) or **1.5–2.5 minutes** (live), while Qwen3-ASR is slower but handles more languages. **AnyDecisionModel** provides typed yes/no, multiple-choice, or rating-scale decisions via local models (MLX, Core AI, or Ollama) or cloud services (Jev). Unlike traditional language-model prompts, it returns probabilities and confidence scores for each option, enabling structured decision-making without text parsing. The package supports models like Clef-Flash (9B parameters) for joint schema decisions or smaller models like Qwen3-4B for emoji classification (e.g., filtering emojis by user input). Both packages require **Swift 6.2+** and **macOS 26/iOS 26+**, with Apple silicon needed for Parakeet/Sortformer. The code is open-source on GitHub, with models hosted on Hugging Face Hub and downloads triggered on-demand by the app.

## Key points

- AnySpeechRecognizer adds Parakeet (25 languages) and Qwen3-ASR (30+ languages) to Apple’s SpeechAnalyzer API, with speaker diarization via Sortformer
- AnyDecisionModel enables typed decisions (e.g., urgency, team assignment) with probabilities and confidence scores, using local models like MLX or cloud services like Jev
- Both packages run on Apple silicon with Core ML/MLX backends; Parakeet transcribes audio in **20–40 seconds per hour** (file) or **1.5–2.5 minutes live**

## Why it matters

These packages bridge Apple’s platform APIs with open models, giving developers flexibility to choose engines for speech recognition and decision-making without vendor lock-in. For Apple developers, it means better multilingual support and structured decision tools without relying solely on Apple’s proprietary models.

## Sources

1. [Introducing AnySpeechRecognizer and AnyDecisionModel: Open Models for Speech and Decisions on Apple Platforms](https://huggingface.co/blog/anydecisionmodel) (Hugging Face, 2026-10-29, primary source)

## Cite

Digest AI, "Hugging Face releases AnySpeechRecognizer and AnyDecisionModel for Apple platforms", 29 October 2026, https://digestai.news/story/hugging-face-releases-anyspeechrecognizer-and-anydecisionmodel-for-app

---

Written by Digest AI's editorial model from the linked sources; the sources are the record. Headlines, digests and key points are written by Digest AI and may be quoted with a link to the story page. Linked articles belong to their publishers. Terms: https://digestai.news/terms#reuse
JSON: https://digestai.news/story/hugging-face-releases-anyspeechrecognizer-and-anydecisionmodel-for-app.json