Hns
hns is a speech-to-text CLI that transcribes your microphone locally with Whisper and copies the text straight to your clipboard.
If your work already lives in a terminal, hns is close to a no-brainer: it is free, MIT licensed, installs in about 30 seconds, and turns every CLI tool into a voice target through stdout piping. The documented `claude "$(hns)"` pattern alone justifies the install for anyone driving coding agents. It is genuinely thin compared with what you get from cloud STT — hns does transcription and clipboard, and you bring your own LLM (Ollama's gemma3:1b, or a hosted model like gpt-4.1-nano) for cleanup and translation. Choose it over Google Speech-to-Text or Azure Speech when privacy and cost matter more than diarization and streaming. If you want a GUI instead, MacWhisper is the more comfortable
Verified 3d ago · liveness 67/100 · cite: rightaichoice.com/tools/hns
- Developers driving AI coding agents from the terminal
- Privacy-conscious users who want fully offline transcription
- Terminal users who build workflows with pipes and stdout
- Writers and note-takers who would rather speak than type
- Anyone who needs a graphical interface or mobile app
- Meetings and multi-speaker recordings that need speaker separation
- Teams wanting a centrally managed, non-technical dictation service
We scan live Reddit threads, YouTube comments, X posts, G2 reviews and other communities — and hand you an honest verdict in under a minute.
- Honest verdict, not marketing
- Real pros & cons from real users
- Attributed quotes with receipts
3 free scans · no card needed
Skip hns if you need a click-to-dictate desktop or mobile app, multi-speaker meeting transcription, or live captioning while you speak — it is a single-user terminal pipe for raw transcription.
hns itself is free, but the first run downloads a Whisper model, which costs bandwidth and disk space before you get any transcription.
hns is a free, MIT-licensed CLI with no subscription, which puts it at the opposite end of the market from per-minute cloud speech APIs like Google Speech-to-Text or Azure Speech. The realistic comparison is a one-time hardware and setup cost against a recurring per-minute bill: solo developers and small teams get the most from hns, while larger organizations that need managed quotas, support contracts, and centralized admin are better served by a commercial provider.
In short
Hns — hns is a speech-to-text CLI that transcribes your microphone locally with Whisper and copies the text straight to your clipboard. Best for Developers driving AI coding agents from the terminal, Privacy-conscious users who want fully offline transcription, Terminal users who build workflows with pipes and stdout. Free to use.
What people actually say about Hns — is it worth it?
We scanned public community sources for Hns on Jul 5, 2026 and could not establish that the discussion we found is about this tool rather than something else sharing its name. Our own analysis of that scan says the posts were off-subject. Rather than publish a sentiment score built on the wrong subject, we publish nothing here and re-run the scan.
Viability Score
How well maintained and how widely used is Hns? Built from what the vendor actually publishes (docs, changelog, tutorials, integrations, pricing), whether the site is live, and how much real users discuss it. How we calculate this
Last calculated: October 2026
How we score →Key Features
- 100% local speech-to-text using Whisper
- Microphone capture with transcription copied to clipboard automatically
- Transcription written to stdout, status messages to stderr for piping
- Works offline after the initial Whisper model download
- Multi-language transcription covering languages Whisper supports
- Install via `uv tool install hns`, pipx, or pip
- Runs on Linux, macOS, and Windows
- Model and language selection via environment variables
- Drive AI coding agents (Claude Code, Codex CLI, Cursor CLI) with dictated prompts
- Append dated journal entries to markdown files such as an Obsidian vault
- Clean up transcription through a local LLM, e.g. `ollama run gemma3:1b`
- Translate dictated speech via an LLM prompt, e.g. polite Japanese with gpt-4.1-nano
- Composable with any CLI tool that reads stdin
- Open source under the MIT license
- Interactive recording flow: press Enter to stop and transcribe
About Hns
hns is a command-line speech-to-text tool for developers who work in the terminal. It captures your microphone, transcribes locally with Whisper, and puts the result on your clipboard — or writes it to stdout so you can pipe it into anything. Because transcription happens entirely on your machine, audio never goes to a cloud API and there is no recurring subscription. Install is one command, `uv tool install hns`, with pipx and pip as alternatives on Linux, macOS, and Windows; the project's quick start promises you are running in 30 seconds after the initial Whisper model download. The design follows the Unix philosophy: progress messages go to stderr, the transcription goes to stdout, so any CLI tool becomes a voice target. The documented patterns show the range — `claude "$(hns)"` to dictate context-rich prompts to Claude Code, Codex CLI, or Cursor CLI; `hns >> "~/obsidian-vault/$(date '+%Y-%m-%d').md"` to append spoken journal entries to a dated note; piping raw transcription through a local LLM like gemma3:1b via Ollama to strip filler words and fix punctuation; and passing it through a model like gpt-4.1-nano to turn dictated English into polite Japanese. Multi-language transcription follows whatever Whisper supports, and models and languages are swapped with environment variables. It is MIT licensed and reads best as a complement to, not a replacement for, a full cloud speech stack.
Behind the Verdict
hns does one job and does it in a way that fits its audience. Install with `uv tool install hns` (or pipx, or pip), run `hns`, see 🎤 Recording..., press Enter, and the transcription lands on your clipboard and in your terminal. That is the whole core loop, and it is why the tool is easy to trust. The strengths are structural. Audio never leaves the machine because recognition runs through local Whisper, so there is no per-minute bill and no third party holding your recordings. After the initial model download it works offline. Because transcription goes to stdout while status messages go to stderr, it composes with everything: the docs' own examples pipe hns into Claude Code, Codex CLI, Cursor CLI, Ollama, and an `llm` CLI call for translation. Environment variables let you pick models and languages without editing config files, which matters if you move between machines or languages. The weaknesses are equally structural, and the docs are upfront about them. hns is a text-in-text-out pipe, not a speech platform: the value stops at raw transcription, so grammar cleanup and translation are workflows you assemble yourself with a local or remote LLM. It is strictly terminal — no GUI, no mobile app, no browser extension. It assumes you can install Python tooling and tolerate a model download before the first offline run. And it is a single-user utility, not something with admin consoles or centralized policy. Where it fits: developers who already drive coding agents from the shell and want faster, richer prompts; note-takers who prefer speaking a journal entry to typing it; anyone with sensitive audio who wants dictation without a cloud account. Where it does not: meetings with multiple speakers, live captioning, or any team that needs a managed service to hand out to non-technical staff. Compared with MacWhisper or VoiceInk, hns trades a polished interface for composability and a zero price tag; compared with cloud STT APIs, it trades managed extras for local control.
Researching Hns? Get your full AI stack in 60 seconds.
Free, no signup — tell us your goal and get tools matched to your budget & existing stack.
Real-world workflow fit
Concrete scenarios for the personas Hns actually fits — and what changes day-one when you adopt it.
You run `claude "$(hns)"`, speak a long, context-rich prompt describing the change you want, press Enter, and Claude Code receives the full instruction while the same text sits on your clipboard for reuse.
Outcome: You write more detailed prompts than you would type, and the agent gets the context it needs on the first pass.
Each evening you run `hns >> "~/obsidian-vault/$(date '+%Y-%m-%d').md"` and speak freely for a few minutes instead of facing a blank page.
Outcome: A dated note builds itself from your spoken thoughts, with no typing and no cloud service holding the audio.
You dictate into hns and pipe the transcription into `llm --model gpt-4.1-nano --system "Translate into polite natural Japanese."`, staying in the terminal the whole time.
Outcome: You send polite, natural Japanese messages at speaking speed without switching to a translation website.
Use Cases
- Dictate detailed prompts to Claude Code, Codex CLI, or Cursor CLI instead of typing them
- Speak journal entries that append to a date-stamped note in an Obsidian vault
- Dictate emails, Slack messages, and documents, then polish the raw text through an LLM
- Translate spoken English into polite Japanese by piping hns into a model like gpt-4.1-nano
- Chain hns into any CLI tool that accepts stdin for custom voice-to-text workflows
- Use offline dictation on a machine where audio must not leave the device
Models Under the Hood
as of 2026-10-18
Limitations
- hns is terminal-only: you install it with uv, pipx, or pip, run it from a shell, and there is no GUI or mobile app.
- Recognition uses local Whisper, so the first run requires a model download and transcription quality tracks whatever Whisper models you point it at.
- Everything after that works offline. hns produces raw transcription and stops there — grammar cleanup, filler-word removal, and translation are pipelines you build yourself by piping the output into a local LLM (the docs use `ollama run gemma3:1b`) or a hosted model such as gpt-4.1-nano.
- It is a single-user tool: no admin console, no shared policy, no centralized rollout for a whole company.
as of 2026-10-08
Verification history
We have re-verified Hns 10 times since . Each pass re-reads the vendor's own pages and re-checks every listed field against that evidence; passes where nothing had changed are marked as such.
- — re-checked, vendor evidence unchanged
- — re-checked, vendor evidence unchanged
- — re-checked, vendor evidence unchanged
- — re-checked, vendor evidence unchanged
- — re-verified summary, description, our verdict, our analysis, pricing model, pricing tiers, features, integrations, who it suits, who should skip it
- — re-verified summary, description, our verdict, our analysis, pricing model, pricing tiers, features, integrations, who it suits, who should skip it
Showing the 6 most recent of 10 verification passes.
Free to cite with attribution — this page re-verifies continuously.
12-month cost
Project the real annual outlay, including the implied monthly cost when only an annual tier is published.
Vendor list price only. Add-on usage, seat overages, and contract minimums are surfaced under Hidden costs & gotchas.
Plans compared
For each published Hns tier: who it actually fits, and what it adds vs. the previous tier. Cross-reference the cost calculator above for projected annual outlay.
Free
$0
Ideal for
Developers, terminal users, and privacy-conscious individuals who want local dictation with no recurring cost.
What this tier adds
Starting tier — the tool is free and MIT licensed: install via uv, pipx, or pip, download a Whisper model, and transcribe locally with no subscription.
Where the pricing makes sense
The company stage and team size where Hns's pricing actually pencils out — and where peers do it cheaper.
hns is a free, MIT-licensed CLI with no subscription, which puts it at the opposite end of the market from per-minute cloud speech APIs like Google Speech-to-Text or Azure Speech. The realistic comparison is a one-time hardware and setup cost against a recurring per-minute bill: solo developers and small teams get the most from hns, while larger organizations that need managed quotas, support contracts, and centralized admin are better served by a commercial provider.
Setup time & first value
How long it actually takes to get something useful out of Hns — broken out by persona, not the marketing-page minute.
Install and first transcription take about 30 seconds once your machine has uv, pipx, or pip — the docs walk through a single command and a test sentence. The variable is the Whisper model download, which can add minutes on a slow connection before the first offline run; after that, starting hns and speaking is immediate.
Switching to or from Hns
How to bring data in from common predecessors and how to get it back out — written for the switcher, not the buyer.
- →From typing prompts into coding agents: keep the same workflow but replace the keyboard with `claude "$(hns)"`.
- →From cloud speech-to-text APIs: point hns at your microphone and route stdout into whatever pipeline previously received API responses.
- →From a GUI dictation app: install hns with uv or pipx and use the clipboard output wherever you pasted before.
- →From manual journaling: add `hns >> "~/obsidian-vault/$(date '+%Y-%m-%d').md"` as your daily note command.
- ↗To MacWhisper: move to the GUI if you want a desktop app with a Mac-native interface.
- ↗To a cloud STT API such as Google Speech-to-Text or Azure Speech: switch when you need managed infrastructure and integrations beyond the terminal.
- ↗To a general LLM voice mode: use it when you want dictation bundled with a chat assistant instead of a standalone CLI.
Integrations
Resources & Guides
- Documentationhns-cli.dev
Docs · Hns
Full product docs from hns-cli.dev
- Quickstarthns-cli.dev
Quick Start · Hns
Get up and running fast from hns-cli.dev
- Documentationhns-cli.dev
Installation · Hns
Full product docs from hns-cli.dev
- Documentationhns-cli.dev
Configuration · Hns
Full product docs from hns-cli.dev
- Documentationhns-cli.dev
Cli Reference · Hns
Full product docs from hns-cli.dev
Tutorials & Learning
YouTube returned 6 videos for “Hns”, and we withheld 6: 6 could not be judged, because “Hns” is a single word that other videos use for other things. We are showing none, because we could not prove any of them are about Hns.
Official links
Tools that pair well with Hns
Common stack mates teams adopt alongside Hns, with the specific reason each pairing earns its keep.
MacWhisper
MacWhisper transcribes audio to text offline on your Mac with Whisper, Parakeet and Qwen models — no audio leaves your machine.
Harker
Harker is a free macOS voice-to-text app that transcribes speech on-device and pastes it into whatever app you're already typing in.
Soniox
Soniox speech-to-text API delivers real-time STT, TTS, and translation in 60+ languages through one unified developer platform.
Featured Head-to-Head Comparisons
Hns vs Spider Cloud
Hns and Spider Cloud serve completely different needs. Hns is a free, open-source offline CLI for speech-to-text, ideal for developers who want to control AI agents by voice while keeping data local. Spider Cloud is a high-throughput web scraping API built for AI agents and RAG pipelines, offering structured output and browser control at low cost. Choose Hns if you need local voice transcription; choose Spider Cloud if you need to feed your agents real-time web data.
Hns vs Temporal Ai
Temporal AI is for teams needing durable, fault-tolerant orchestration of AI agents and workflows with cloud or self-hosted options; Hns is a lightweight, free CLI for local speech-to-text, ideal for voice programming with AI coding agents. Choose Temporal if you need reliability and state persistence for complex workflows; choose Hns if you just want a simple voice-to-clipboard tool for developer productivity.
Hns vs Voyage Ai
Choose Voyage AI if you need enterprise-grade, domain-optimized embeddings for RAG on sensitive or specialized documents—especially in finance, legal, or code. Choose hns if you're a developer who wants a dead simple, offline voice-to-text CLI tool to pipe into AI coding agents like Claude Code or for private note-taking. They serve entirely different needs; your decision hinges on whether your problem is search accuracy or voice input.
Alternatives to Hns
View all
MacWhisper
MacWhisper transcribes audio to text offline on your Mac with Whisper, Parakeet and Qwen models — no audio leaves your machine.
Frequently Asked Questions
Best-of guides
Used Hns? Help shape our editorial sentiment research.