TypeAloud vs Spokenly
A Spokenly alternative
that is open source.
Spokenly's local transcription is free and unlimited, with no account, and so is TypeAloud's. The real differences are the speech model, whether you can read the source, and how much AI sits on top of the transcript.
TypeAloud and Spokenly compared
| TypeAloud | Spokenly | |
|---|---|---|
| Price | Free | Free for local models and your own API keys; Pro at $9.99/month or $99.99/year |
| Local speech model | Qwen3-ASR 0.6B, or 1.7B as an option | Whisper and Parakeet models |
| Word limit | None with a local model | None with a local model |
| Account | Not needed | Not needed for local use |
| Source code | Open source (GPL-3.0) | Closed source |
| AI text processing | No. It writes what you said | Yes, with GPT, Claude or Gemini |
| Platforms | macOS; Windows and Linux builds are experimental | macOS, Windows, Linux and iPhone |
Competitor details checked on 18 Oct 2026 against spokenly.app. Prices and limits change; the official pages are authoritative.
Chinese and English in one sentence
Spokenly runs Whisper and Parakeet on your device. Parakeet's multilingual model covers European languages, not Chinese, and Whisper picks one language per recording.
TypeAloud's local model, Qwen3-ASR, works out the language from the audio. That suits the way a lot of technical work in Chinese sounds: Chinese grammar around English product names, API names and half the nouns, all in one breath.
Source you can read
TypeAloud is GPL-3.0 on GitHub. When it says local mode keeps your audio on your Mac, you can check the code that records it, transcribes it and saves it to History.
It also does one job: dictation into the app you are using. There is no Pro plan, so there is nothing to upsell inside the app.
Where Spokenly is the better choice
Spokenly can pass your transcript through GPT, Claude or Gemini with modes and instructions, run voice-controlled actions on macOS, record meetings and work as an iPhone keyboard. TypeAloud does none of these.
If you dictate mainly in English and want AI processing, an iPhone keyboard or meeting notes, Spokenly fits better.
Common questions
- Both are free. Why pick TypeAloud?
- Mostly for its local model. If your dictation mixes Chinese and English, Qwen3-ASR handles that without a language setting. TypeAloud is also open source.
- Can I bring my own API key?
- Yes. Cloud transcription is optional and uses your own Groq or OpenAI key, which those providers may charge for.
- Does my audio leave my Mac?
- Not in local mode. The model runs on your device and transcripts are saved there. Audio is only uploaded if you choose a cloud engine with your own API key.
- Which Macs can run it?
- Apple Silicon is recommended for local dictation. The recommended model is a one-time download of about 1 GB.
Local transcription. No account. No subscription.
Download for Mac