“There might be a simpler way to solve this.”
Take out your phone, record the thought, then carry on.
A voice note for the thought you want to keep.
Tap record on your phone. Turn a complete thought into text and a useful note, then revisit it in SpeakPen, Obsidian or your preferred AI.
Free plan: 3 recordings per month, up to 5 minutes each.
iOS · Android
Your scattered thoughts begin to connect.
“People want to walk more, but do not know where to go. What about a small product that suggests a quiet route each day?”
Original words · time in recording · 08:42
“A route you can finish in ten minutes might matter most. No planning needed. Just step outside.”
Original words · time in recording · 13:16
“Write up three routes I already know and ask two friends to try them. Find out what would actually get them outside.”
AI conversation · action draft · 20:35
On a walk, between errands, or just after a shower. Say it while it is fresh and carry on with your day. Return later to a thought you can still develop.
“There might be a simpler way to solve this.”
Take out your phone, record the thought, then carry on.
“Wait, there is another way to approach that idea.”
Open your phone and add the new clue.
“These separate thoughts are about the same thing.”
Revisit what matters and decide what to continue.
One deliberate recording is enough. Start when you have something to say and stop when you are done.
Download SpeakPeniOS or Android. Capture a thought, or talk through a fuller idea.
Download SpeakPen →On-device transcription is used when available. Otherwise, cloud transcription takes over. Your original audio stays available to revisit.
Download SpeakPen →No model setup or continuous recording. Processing can take time; offline recordings wait on your phone.
Keep the original clues. Choose which ones to pursue and develop them with the tools you already use.
Capture · now
Say what you are thinking. The title and structure can come later.
Revisit · later
Read the summary, revisit the original words, and listen to the audio when you want to check a detail.
Build · every day
Keep notes in SpeakPen, or configure the plugin to bring them into your own Obsidian vault.
Develop · next
Connect an AI assistant or share selected notes. Discuss options and choose your next action.
Let ChatGPT, Claude or any MCP agent read your notes directly — or point it at your vault
An illustrative idea journey. Review AI suggestions before acting on them.
Accumulate and revisit your daily notes in SpeakPen. Add Obsidian if it fits your habits, and connect AI when you want to explore a question.
Configure the SpeakPen Sync plugin to bring voice notes into your own vault. Use your existing Obsidian sync setup across your devices.
Setup steps →ChatGPT, Claude and other AI assistants can read your notes — read-only, never your audio
Connect your AI →Connect to my SpeakPen notes: read https://speakpen.app/connect and set it up.See how voice notes sync to Obsidian →Which Obsidian transcription plugins still work →
Your observations, preferences and questions are a useful starting point. Keep them, then ask AI to find connections, ask better questions and explore next steps.
Connect your AI →“Bring together my ideas about walking this week. What could become a small product? Give me a plan I could test this weekend.”
Test whether people want it before deciding to build an app.
Illustrative prompt and response · external AI workflow, not a live result
Start free. Upgrade when you need more.
No manual model selection or download is required. SpeakPen first tries available local transcription and uses the cloud if it cannot run. Audio and text sync to your account; offline recordings are saved for later processing.
SpeakPen turns your recordings into text and summaries you can revisit, search, and organize. You can use it on its own. If you use Obsidian or an AI assistant, connect them when you need them.
After you record or import audio, SpeakPen processes it into text and a summary. Automatic language detection supports 99 languages.
Yes. Install "SpeakPen Sync" from Obsidian's community plugin browser, then paste an API token — generate one in the SpeakPen web app at speakpen.app/app, under Settings → Connections (Obsidian card → Generate token). New notes then arrive in your vault as Markdown, each with a title, summary, and full transcript.
Obsidian's own Quick Capture and most inbox apps are built for a typed line. Voice plugins for Obsidian only record while Obsidian is open, and Mac transcription apps need your Mac. SpeakPen is for the ideas you would rather talk through: it records on your phone or in a browser, lets you pause and ramble, detects the language, and syncs a titled, summarized Markdown note into your vault afterwards.
Automatic language detection covers 99 languages — just speak and SpeakPen figures it out. If you prefer to pin a language manually instead of relying on detection, the manual picker currently offers 10: English, Chinese (Simplified and Traditional), Japanese, Korean, Spanish, French, German, Portuguese, and Italian.
SpeakPen offers a free plan with 3 recordings (up to 5 minutes each) so you can experience the full pipeline. The Pro plan at $9.99/month unlocks 60 recordings per month with up to 20 minutes each.
Yes, SpeakPen supports importing audio files in addition to recording directly. Imported audio goes through the same AI transcription and summary pipeline as live recordings.
No. Transcription, summaries, action items, rewriting, and asking questions across your notes all happen inside SpeakPen. The agent step exists because your notes are plain Markdown in a folder you control — anything you already point at that folder can read them, so nothing has to be integrated on our side.
Start with one sentence today. Give tomorrow’s you something to continue.
iOS · Android