TLDRocket
Sign in

Shall We Talk

Product Hunt Jobless

Shall We Talk turns speech into text on iPhone and Mac. It keeps your wording, but strips fillers and misheard words.

Based on reporting by Product Hunt, Jobless — read the original for the full story.

Summary, retelling and take written by AI under human oversight; images are AI-generated illustrations. How we work · Report an error

Shall We Talk launched today as a voice dictation tool for iPhone and Mac, with a clear promise: turn speech into clean text without mangling the way you actually talk.

It works in a few different ways. You can dictate into any app through a voice keyboard that types at your cursor, trigger it with Shortcuts, or use a Mac hotkey. It also handles meetings, producing speaker-labeled transcripts and structured summaries.

The cleanup pass is unusual in a good way. The app acts like a proofreader, not a rewrite machine. It removes fillers and misheard words, but keeps your wording, your tone, and even Chinese-English code-switching intact.

Shall We Talk is open source under GPL-3.0, and it asks users to bring their own speech and LLM API keys. Product Hunt says the project advanced from build 200 to 243 with GPT-6 Astra, which suggests the team is still pushing on the model side while keeping the core product fairly simple.

My take — AI-written commentary, not fact-checked reporting

Open source plus bring-your-own-keys is the least glamorous way to build AI software, which is exactly why it ages better. The industry keeps pretending the magic is the model; most users just want their words back without the clown makeup.

Read more about this at: Product Hunt

Related stories

The daily briefing

Every AI story that matters, in your inbox by 8am.

TLDRocket reads all relevant sources, removes duplicate coverage, and summarises the day in two minutes. Follow companies and topics for alerts, or get the briefing in Slack. Free, no spam, unsubscribe anytime.