Free tools Windows power users keep installed
One-click scans. No signup required.
Some links on this page are affiliate links: if you buy through them we may earn a commission, at no extra cost to you.
Ollama can now turn text into speech locally on Apple Silicon Macs, starting with Phonon, a family of models from Gradium. You can speak text from the terminal, choose a voice, pipe in a chat model’s answer, save audio as a WAV file, or use an OpenAI-compatible API. The announcement is dated October 24, 2026; its launch post and current capability documentation give different minimum Ollama versions, so check the official speech documentation before installing or upgrading.
What does Ollama’s new speech support do?
It adds local text-to-speech (TTS) to Ollama on Apple Silicon Macs. The first supported family is Phonon, developed by Gradium. Rather than send text to a hosted speech service, Ollama generates speech on the Mac; the resulting audio can play as it is generated or be saved for later.
As an Amazon Associate I earn from qualifying purchases.
The October 24, 2026 announcement lists English, French, German, Spanish, Portuguese, and a multilingual model covering all five. At the time, Ollama said Windows and Linux support was coming soon. That was a time-specific statement, not confirmation of their current availability; check the current documentation for supported platforms.
What do you need to get started?
You need Ollama installed on an Apple Silicon Mac and a speech model pulled locally. The launch announcement says to use Ollama 0.40.11 or later, while the linked speech documentation says 0.40.9 or later. Because those official sources disagree, do not treat either as a settled current minimum: check the documentation and install the version it currently specifies. Ollama’s download page provides installation options and notes that local-model speed depends on hardware. The supplied information does not establish a minimum memory requirement.
Popular on Amazon
- Magic Keyboard is available with Touch ID, providing fast, easy and secure authentication for logins and to unlock your Mac.
- Magic Keyboard with Touch ID and Numeric Keypad delivers a remarkably comfortable and precise typing experience.
- It features an extended layout, with document navigation controls for quick scrolling and full-size arrow keys, which are great for gaming.
- The numeric keypad is also ideal for spreadsheets and finance applications.
- It’s wireless and features a rechargeable battery that will power your keyboard for about a month or more between charges.
To install the English model and generate a short spoken phrase:
ollama pull phonon
ollama run phonon "Hello from Ollama."Ollama can play the generated audio through the computer’s speakers. You do not need a separate audio accessory to use the feature, though you can save the result as a file instead.
Quick wins for a faster PC:
Scan for outdated or missing drivers - takes under a minuteDriver Scan →Clear out junk files and repair common Windows errorsFree Scan →Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →How do you choose a language and voice?
The English model is called phonon. The announcement names four single-language 40M models and one five-language model:
| Model | Language coverage | License stated in announcement |
|---|---|---|
phonon |
English | Apache 2.0 |
phonon:40m-fr |
French | CC BY-NC 4.0 |
phonon:40m-de |
German | CC BY-NC 4.0 |
phonon:40m-es |
Spanish | CC BY-NC 4.0 |
phonon:40m-pt |
Portuguese | CC BY-NC 4.0 |
phonon:40m-multilingual |
English, French, German, Spanish, and Portuguese | CC BY-NC 4.0 |
The announcement says the English model uses Apache 2.0, while the 40M language models use CC BY-NC 4.0, which restricts use to noncommercial purposes. Check the applicable model license before using generated speech, especially in a commercial project.
Also popular on Amazon
- Magic Keyboard is available with Touch ID, providing fast, easy and secure authentication for logins and to unlock your Mac.
- Magic Keyboard with Touch ID and Numeric Keypad delivers a remarkably comfortable and precise typing experience.
- It features an extended layout, with document navigation controls for quick scrolling and full-size arrow keys, which are great for gaming.
- The numeric keypad is also ideal for spreadsheets and finance applications.
- It’s wireless and features a rechargeable battery that will power your keyboard for about a month or more between charges.
To choose an English voice, pass its name with --voice. For example:
ollama run phonon --voice Hadley "Hello from Ollama."Run ollama show phonon to see the voices available for that model. The documentation also names Zoey, Easton, and Freya among its English voices. The voice catalog is from Gradium; the supplied sources do not establish a quality ranking among voices or language variants.
How can you speak piped text or a chat response?
Phonon can read text supplied through a shell pipe, including a response generated by another Ollama model. This is useful when you want spoken output without copying text between applications. Ollama says audio is streamed as it is generated, so playback can begin before the full response is complete.
The launch announcement documents piping a chat model’s answer into Phonon. The exact shell command depends on the chat model and how you invoke it; consult the announcement examples for the documented pattern. You can also redirect speech output to a file to create WAV audio.
Also popular on Amazon
- Stable 2.4GHz Wireless Mouse:YUNZII C2 mouse come with a 2.4G receiver stored in the gaming mouse. Install the battery, insert the receiver into your device, like computers, laptops, tablets and more. Turn on the switch, and enjoy instant smooth responsiveness without any complicated setup required. Enjoy seamless cable-free connectivity with a reliable 2.4GHz USB receiver of this cute mouse
- 3-Level DPI Adjustment for Precision: With this gaming wireless mouse, you can effortlessly switch between 800, 1200, and 1600 DPI settings to match different needs. Ideal for detailed design work (800 DPI), office browsing (1200 DPI), or fast-paced gaming (1600 DPI). The 250Hz polling rate ensures accurate tracking and smooth performance
- Portable Silent Mouse: Compared to office or normal gaming mouse, YUNZII C2 noiseless mouse produces a maximum click noise of just 45dB. Low-noise but responsive buttons minimize disruption, making it optimal for offices, libraries, cafes, or late-night gaming sessions. The portable mouse 's compact, lightweight design makes it easy to carry in your bag, offering quiet yet responsive clicks wherever you go
- Ergonomic Silicone Mouse: This computer mouse is ergonomically designed to comfortably fit the palm of your hand. It's crafted from a soft, food-grade memory silicone that is especially skin-friendly. This material offers personalized support that reduces pressure and fatigue, while its sculpted, compact form ensures a natural and enveloping grip—perfect for long hours of work or gaming
- Long-Lasting Battery & Broad Compatibility: Engineered with advanced low-power technology, this portable mouse runs on a single AA battery (included) to deliver up to 6 months of uninterrupted use—ideal for both extended gaming sessions and daily office tasks.This cordless mouse supports stable 2.4GHz plug-and-play connectivity and offers broad compatibility with macOS, Windows, PCs, laptops, and more, giving you smooth performance across all your devices
Can an app call Ollama’s speech API?
Yes. Ollama provides an OpenAI-compatible /v1/audio/speech endpoint, so an application can request generated speech from a local Ollama instance instead of sending the text to a hosted speech API. The capability documentation describes using the endpoint, including through the OpenAI SDK with Ollama configured as the local base URL.
API input is limited to 4,096 characters per request. The CLI handles longer text by splitting it at paragraph and sentence breaks. Audio defaults to mono 24 kHz WAV; raw 16-bit PCM is also documented. For predictable pronunciation and formatting, note that numbers, prices, symbols, email addresses, and URLs are spoken after normalization, while emoji, bullets, and Markdown markers are skipped.
Unsupported options and unknown voice names return a 400 error. Speech models are for speech generation: chat, completion, and embedding requests sent to them also return a 400 error. Use a chat model to generate or answer text, then route that text to a speech model when you want audio.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.How fast and accurate is it?
Ollama reports that its implementation generated speech 35% faster than Gradium’s ptts runtime in a comparison on a MacBook Pro with M5 Max. The company says the test generated 45 seconds of speech and used the median of 10 runs. It also describes different numerical precision and hardware paths: Gradium’s runtime used 8-bit weights on the CPU, fp16 on the Neural Engine, and f32 on the GPU, while Ollama used bf16.
Also popular on Amazon
![Apple Magic Keyboard with Touch ID for Mac Models with Apple Silicon [Lightning Port] (QWERTY English) Silver (Renewed)](https://m.media-amazon.com/images/I/41PIzgJdm6L._SL160_.jpg)
- WIRELESS, RECHARGEABLE CONVENIENCE - Magic Keyboard with Touch ID connects wirelessly to your Mac via Bluetooth. And the rechargeable internal battery means no loose batteries to replace.
- WORKS WITH ANY MAC WITH APPLE SILICON - It pairs automatically with your Mac with Apple silicon so you can get to work right away. See the list of compatible devices above. Requires a Mac with Apple silicon using macOS 11.4 or later.
- ENHANCED TYPING EXPERIENCE - Magic Keyboard delivers a remarkably comfortable and precise typing experience.
- QUICK UNLOCK WITH TOUCH ID - Touch ID gives you a fast, easy, secure way to unlock your Mac and sign in to apps and sites.
- GO WEEKS WITHOUT CHARGING - The incredibly long-lasting internal battery will power your keyboard for about a month or more between charges. (Battery life varies by use.) Comes with a woven USB-C to Lightning Cable that lets you pair and charge by connecting to a USB-C port on your Mac.
That is a vendor-reported comparison under one described setup, not an independent benchmark or a guarantee for other Macs. Hardware differs, and Ollama itself says local-model speed depends on the machine. The announcement also says Ollama matched the word error rate of the Hugging Face Transformers reference implementation, based on 200 English sentences from the Seed-TTS test set transcribed with Whisper large-v3. That accuracy result is also Ollama’s claim, not independent verification.
What should you check before using Phonon?
- Platform: The announcement specifies Apple Silicon Macs. It does not establish support for every Mac or a minimum memory requirement.
- Version: The launch post says 0.40.11 or later, but the linked capability documentation says 0.40.9 or later. Follow the current official instructions rather than assuming the two statements match.
- License: The English
phononmodel is listed as Apache 2.0; the 40M language models are listed as CC BY-NC 4.0 and are noncommercial. - Input formatting: API requests allow up to 4,096 characters. Markdown markers, bullets, and emoji are skipped, while several kinds of numbers and symbols are normalized for speech.
- Other platforms: Windows and Linux were described as coming soon in the October 2026 announcement. Verify current availability in the documentation rather than relying on that dated wording.
Frequently Asked Questions
Does Ollama text-to-speech work without an internet connection?
The announced feature runs speech models locally on Apple Silicon Macs, rather than relying on a hosted speech-generation service. The supplied documentation does not specify whether every installation or setup step can be completed offline.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
Can Phonon read a PDF or document directly?
The documented interface accepts text, not a PDF or document file as such. Extract the text first, then pass it to the CLI or API; formatting such as bullets and Markdown markers may be skipped.
Can I use Phonon-generated audio in a commercial product?
The announcement lists Apache 2.0 for the English model and CC BY-NC 4.0 for the 40M models. Because the latter is noncommercial, check the exact model’s license and terms for your intended use before shipping audio commercially.
