---
title: "Speech recognition guides"
description: "Guides to choosing, running and measuring speech-to-text models: comparisons, costs, on-device and self-hosted setups, benchmarks and methodology."
canonical: "https://www.fermionresearch.com/speech/"
source: "Fermion Research"
---

# Speech recognition guides

How to choose, run and measure speech-to-text models, with dated sources. Every number on these pages is measured or quoted from the vendor's own page.

## Choosing a model

- [The best open-source speech-to-text model for your use](https://www.fermionresearch.com/speech/best-open-source-speech-to-text-model/)Open speech models ranked on the seven Open ASR Leaderboard English sets, with size and languages. Which is best for accuracy, small size, or many languages.
- [The best way to run speech-to-text on Apple silicon](https://www.fermionresearch.com/speech/best-speech-to-text-for-apple-silicon/)MLX, Core ML on the Neural Engine, whisper.cpp, WhisperKit, FluidAudio and Apple's SpeechAnalyzer compared: which engine and model to use on an M-series Mac.
- [Whisper alternatives: open models, faster runtimes and APIs](https://www.fermionresearch.com/speech/whisper-alternatives/)Alternatives to OpenAI Whisper by need: lower English error, smaller downloads, faster Whisper runtimes or hosted APIs. Measured numbers, and when Whisper wins.
- [On-device speech-to-text for iOS, macOS, Android and desktop apps](https://www.fermionresearch.com/speech/on-device-speech-to-text-for-apps/)How to add on-device speech recognition to an iOS, macOS, Android or desktop app: model sizes, runtimes such as Core ML, languages and the privacy trade-offs.
- [Custom vocabulary in speech recognition: hotwords, keyterms and prompts](https://www.fermionresearch.com/speech/custom-vocabulary-speech-recognition/)How custom vocabulary works in speech-to-text: hotwords, keyterms and prompts, with syntax and limits for Phonon, Deepgram, AssemblyAI, OpenAI and Whisper.
- [How much does a speech-to-text API cost per hour?](https://www.fermionresearch.com/speech/speech-to-text-api-cost/)Dated list prices per audio hour for speech-to-text APIs, from about $0.01 for hosted Whisper turbo to $0.96, and what self-hosting an open model costs.
- [Self-hosted vs API speech-to-text: how to decide](https://www.fermionresearch.com/speech/self-hosted-vs-api-speech-to-text/)When to run an open speech-to-text model yourself and when to pay a hosted API. Cost at your volume, latency, privacy, languages and features, side by side.
- [Multilingual speech recognition for European languages](https://www.fermionresearch.com/speech/multilingual-speech-recognition-european-languages/)Word error by language on FLEURS for 25 European languages, from Spanish to Maltese, for three open models, and which open models cover which languages.
- [Offline speech recognition in Python](https://www.fermionresearch.com/speech/offline-speech-recognition-python/)Transcribe audio in Python without a cloud API: fermion (Phonon), openai-whisper, faster-whisper, Vosk and Moonshine compared, with install and run commands.

## Model comparisons

- [Phonon-2.5 vs Whisper](https://www.fermionresearch.com/models/phonon-2-5/vs-whisper/)Phonon-2.5 is a 156 MB open speech model for 25 languages. How it compares with OpenAI Whisper on FLEURS word error, download size, languages and where it runs.
- [Phonon-2.5 vs Parakeet TDT 0.6B v3 and Parakeet Redux](https://www.fermionresearch.com/models/phonon-2-5/vs-parakeet/)Per-language FLEURS word error for Phonon-2.5 (156 MB), Parakeet Redux (178 MB) and Parakeet TDT 0.6B v3 (2.5 GB), all scored with one code on 25 languages.
- [Open speech-to-text models under 200 MB, compared](https://www.fermionresearch.com/models/phonon-2-5/small-models/)Five open speech recognition models between 156 and 178 MB compared on the seven Open ASR Leaderboard English sets, plus FLEURS for the two that cover 25 languages. Measured, dated.
- [Self-host a speech-to-text server with an OpenAI-compatible endpoint](https://www.fermionresearch.com/models/phonon-2-5/self-host/)Run a 156 MB open speech model as a transcription server on CPU, Apple silicon or NVIDIA GPU. One command exposes POST /v1/audio/transcriptions; Docker images for CPU and CUDA.
- [Speech recognition on the Apple Neural Engine with Core ML](https://www.fermionresearch.com/models/phonon-2/core-ml/)Phonon-2 Core ML runs a 5.21 % word error open speech model on the Apple Neural Engine at 606 times realtime on an M5 MacBook Air, with Swift and Python runners. How to run it.
- [The Open ASR Leaderboard and the Phonon models](https://www.fermionresearch.com/models/phonon-2-5/open-asr-leaderboard/)What the Hugging Face Open ASR Leaderboard measures, how word error is normalised, and the Phonon rows: Phonon-2.5 and Phonon-2.5 English listed since November 2026.
- [Phonon-2.5 vs Whisper large-v3-turbo](https://www.fermionresearch.com/models/phonon-2-5/vs-whisper-large-v3-turbo/)Phonon-2.5 averages 5.06 % English word error on seven Open ASR sets, Whisper large-v3-turbo 6.58 %, at 156 MB against 1,618 MB. Turbo covers 99 languages.
- [Phonon-2.5 vs Deepgram Nova-3](https://www.fermionresearch.com/models/phonon-2-5/vs-deepgram/)Deepgram Nova-3 is a hosted speech API from $0.0043 a minute with diarization; Phonon-2.5 is a 156 MB open model you run yourself. Prices, features, trade-offs.
- [Phonon-2.5 vs AssemblyAI Universal-3.5 Pro](https://www.fermionresearch.com/models/phonon-2-5/vs-assemblyai/)AssemblyAI Universal-3.5 Pro has lower English word error than Phonon-2.5 (4.46 % vs 5.06 %) at $0.21 an hour; Phonon-2.5 is 156 MB and runs on your hardware.
- [Phonon-2.5 vs ElevenLabs Scribe v2](https://www.fermionresearch.com/models/phonon-2-5/vs-elevenlabs-scribe/)ElevenLabs Scribe v2 beats Phonon-2.5 on English word error on all seven Open ASR sets (4.06 % vs 5.06 %). Phonon-2.5 runs on your hardware with no hourly fee.
- [Phonon-2.5 vs Microsoft MAI-Transcribe-2](https://www.fermionresearch.com/models/phonon-2-5/vs-microsoft-mai-transcribe/)MAI-Transcribe-2 is Microsoft's hosted speech model on Azure: 60 languages, diarization, $0.10 an hour promo. Phonon-2.5 is a 156 MB open model you self-host.
- [Phonon-2.5 vs Meta Muse Voice Transcribe](https://www.fermionresearch.com/models/phonon-2-5/vs-meta-muse/)Muse Voice Transcribe: 4.93 % English word error; Phonon-2.5: 5.06 %; Phonon-2.5 English: 4.82 %. Muse is a $0.18 an hour API; Phonon runs on your hardware.
- [Phonon-2.5 vs Mistral Voxtral](https://www.fermionresearch.com/models/phonon-2-5/vs-voxtral/)Phonon-2.5 averages 5.06 % English word error on seven Open ASR sets; open Voxtral models 5.21 to 6.12 %. Voxtral also answers questions about audio.
- [Phonon-2.5 vs Moonshine](https://www.fermionresearch.com/models/phonon-2-5/vs-moonshine/)Moonshine's streaming models run from 34 M to 245 M parameters on phones, browsers and a Raspberry Pi. Phonon-2.5 is 156 MB with 25 European languages.
- [Phonon-2.5 vs Qwen3-ASR](https://www.fermionresearch.com/models/phonon-2-5/vs-qwen3-asr/)Qwen3-ASR 1.7B has lower English word error than Phonon-2.5 (4.43 % vs 5.06 %); Phonon-2.5 beats Qwen3-ASR 0.6B and covers 9 European languages Qwen does not.
- [Phonon-2.5 vs NVIDIA Canary](https://www.fermionresearch.com/models/phonon-2-5/vs-canary/)Canary-1B-v2 covers Phonon-2.5's 25 languages and translates; Phonon-2.5 has lower English word error (5.06 % vs 5.81 %). Canary-Qwen 2.5B scores 4.55 %.

## Benchmarks and methodology

- [How Fermion Research measures speech models](https://www.fermionresearch.com/speech/benchmark-methodology/)Full public test sets, one scoring code for every model in a table, the leaderboard's English normaliser, diacritics kept for FLEURS, and our rows marked.
- [FLEURS results in 25 languages](https://www.fermionresearch.com/speech/fleurs-results-25-languages/)Phonon-2.5 averages 11.71 % word error on FLEURS across 25 languages, against 12.18 % for Parakeet Redux and 13.26 % for Parakeet TDT 0.6B v3. Every language.
- [English speech recognition benchmark results](https://www.fermionresearch.com/speech/english-benchmark-results/)Phonon-2.5 English averages 4.82 % word error on the seven Open ASR Leaderboard English sets and 5.60 % on the current nine, scored with its code.

## Deployment

- [Replace expensive API model calls with a small model built for the task](https://www.fermionresearch.com/deploy/replace-api-models-with-small-models/)Most production LLM calls do one narrow job. A small model trained for that job runs on your own hardware or on the device, answers in milliseconds, and cuts the per-call cost. How Fermion builds them.
- [On-device and on-premises speech recognition for regulated teams](https://www.fermionresearch.com/deploy/on-device-speech-recognition/)Transcribe and dictate without sending audio to a cloud API. Phonon runs on CPUs, Apple silicon and NVIDIA GPUs inside your environment; Detta dictates on the Mac's Neural Engine. Open weights, measured accuracy.

Source: [https://www.fermionresearch.com/speech/](https://www.fermionresearch.com/speech/)
