--- title: "Best OpenWhispr Alternative in 2026 (Private, No Setup)" description: "The best OpenWhispr alternative in 2026. Yaps, Handy, MacWhisper, SuperWhisper, VoiceInk and Wispr Flow compared on privacy, platforms, setup and upkeep." canonical: "https://www.yaps.ai/blog/openwhispr-alternative" language: "en" --- Entry 99 Comparison · 06 JUL 2026 # Best OpenWhispr Alternative in 2026 (Private, No Setup) OpenWhispr is a genuinely good open-source, on-device dictation app, and it now ships an iOS build alongside desktop. What it still does not have is an Android app. Here are the best OpenWhispr alternatives in 2026, ranked honestly and checked for whether they are still maintained. Yaps Team 16 min read [Back to journal](/blog) **In this entry** ENTRY 99 · 8 SECTIONS - [0.0 Preface](#preface) - [1.0 What Is OpenWhispr?](#what-is-openwhispr) - [2.0 The Best OpenWhispr Alternatives in 2026](#the-best-openwhispr-alternatives-in-2026) - [3.0 Yaps versus OpenWhispr, side by side](#yaps-versus-openwhispr-side-by-side) - [4.0 Who should choose which](#who-should-choose-which) - [5.0 Coming from OpenWhispr? Here is the switch](#coming-from-openwhispr-here-is-the-switch) - [6.0 Frequently Asked Questions](#frequently-asked-questions) - [7.0 Final thoughts](#final-thoughts) - [·1. Yaps: private voice on every device, no setup](#1-yaps-private-voice-on-every-device-no-setup) - [·2. OpenWhispr: the open-source original you are comparing against](#2-openwhispr-the-open-source-original-you-are-comparing-against) - [·3. Handy: the free, no-account open-source peer](#3-handy-the-free-no-account-open-source-peer) - [·4. FluidVoice: the biggest Mac-native open-source option](#4-fluidvoice-the-biggest-mac-native-open-source-option) Entry 99 · The Yaps Journal 06 JUL 2026 · 16 min read Comparison ![Best OpenWhispr Alternative in 2026 (Private, No Setup)](/images/blog/openwhispr-alternative.jpg) 0.0 ## Preface If you are searching for the best OpenWhispr alternative, you are probably one of two people. Either you love what OpenWhispr does but you live on your phone and there is no Android app for it, or you like the idea of local, private dictation and you would rather not clone a GitHub repo and keep a developer tool running to get it. Let us be clear up front. OpenWhispr is a genuinely good project. It is fully open source under an MIT license, it runs transcription on your own device with whisper.cpp and Parakeet, it collects no telemetry, and it is free to run local models forever. If you want an auditable, self-hostable, desktop dictation tool and you are comfortable around a terminal, OpenWhispr is a strong pick and honestly hard to beat on those terms. But "best OpenWhispr alternative" is a real query for a reason. One thing has changed since most comparison pages were written, and we are correcting it here: **OpenWhispr's own site now lists iOS alongside macOS, Windows and Linux**, so the "iOS coming soon" line you will still read elsewhere is out of date. What has not changed is that there is no Android app at all. It is also, by its own description, a power-user and developer-friendly tool. So if you want private voice on an Android phone, or a full voice workspace rather than a dictation utility, or an app that just installs and runs with no setup, you are looking in the right place. Everything below was checked against the live project on 2 September 2026. That matters more in this corner of the market than most: this category is full of indie and open-source projects, several well-known ones have quietly stopped shipping while their websites stayed up, and recommending an abandoned dictation app is worse than recommending nothing. We build Yaps, so we are biased and we will say so plainly. We also think the honest way to earn your trust is to tell you exactly where OpenWhispr wins, where each alternative fits, and where Yaps is the better default. Open-source desktop dictation ### OpenWhispr Fully open-source, on-device dictation and bot-free meeting transcription for macOS, Windows, Linux, and now iOS. No Android app. Markets itself as a developer-friendly, power-user tool. Private voice, no setup ### Yaps On-device dictation plus a full voice workspace: voice notes, Studio audio-file transcription, read-aloud, and vault sync. Ships on Android, Windows, macOS, Linux, and a Chrome extension today. Download and go, no GitHub. Try it yourself You could have dictated this whole article by now. Yaps turns your voice into clean text. On your device, even offline. [Download Yaps](/download)iPhone · Android · Windows · Mac · Linux 1.0 ## What Is OpenWhispr? OpenWhispr is a free, MIT-licensed, open-source dictation app that turns speech into text using models running on your own machine. It runs Whisper locally in five sizes (Tiny at 75 MB, Base at 142 MB, Small at 466 MB, Medium at 1.5 GB, and Turbo at 1.6 GB) plus NVIDIA's Parakeet, or optionally cloud models through your own API keys. It ships builds for macOS, Windows, Linux and iOS, collects no telemetry, stores transcription history in a local SQLite database, and also produces bot-free meeting notes by recording and transcribing locally rather than sending an assistant into your call. There is no Android app. 2.0 ## The Best OpenWhispr Alternatives in 2026 Here is the shortlist, ranked for most people who land on this search. The deeper write-ups follow. Every tool below is real, shipping, and verified against its live release history, and every description is accurate to what each one actually does. ### 1. Yaps: private voice on every device, no setup Yaps is the alternative to reach for if you want the on-device, no-telemetry philosophy that drew you to OpenWhispr, but you also want it on your phone, and you would rather not maintain a developer tool. Start with the thing OpenWhispr cannot match: platform breadth. Yaps ships on Android, iPhone and iPad, Windows, macOS, and Linux, plus a "Save to Yaps" Chrome extension. OpenWhispr is desktop-only and has no Android app. If you dictate on your phone, that difference is the whole story. On Android, Yaps is a full keyboard: tap the dictation button and speak into any app, or type with tap and glide typing, themes, and clipboard history for text and images. Core dictation runs on-device and works offline. The workflow on the desktop is the same shape you already like. You push the Yaps hotkey, you speak, and your words land at the cursor in whatever app you are using. Yaps also cleans up messy dictated text on-device: it strips filler words and self-corrections, fixes punctuation and capitalisation, and formats lists and numbers, all without sending your audio anywhere. Dictation is multilingual, around 25 languages, auto-detected from your speech so you never toggle a language setting. Then Yaps keeps going where a dictation utility stops. Voice notes capture spoken thoughts into text, Kanban boards, and checklists, with local search and export to Markdown and plain text. The Studio editor transcribes imported audio files offline into text or SRT subtitles, so an interview recording becomes a timed transcript on your own machine. Read-aloud gives you 18 voices on desktop and 2 on mobile to hear a draft before you send it. And vault sync moves your notes between phone and desktop over your local network or an encrypted peer-to-peer link. Yaps is built for everyday use rather than developers: a 7-day free trial gives you access to every feature, and paid plans are Pro at $15/mo and Max at $25/mo. If you want the deeper picture, the [dictation feature page](/features/dictation) and our [offline dictation guide](/blog/offline-dictation-guide) walk through how the on-device side works. **Best for:** anyone who wants private, on-device voice on their phone and their computer, with notes, transcription, and read-aloud in one app, and zero setup. **Honest trade-off:** Yaps is not open source, and OpenWhispr is. If auditable code or self-hosting is a hard requirement, that point goes to OpenWhispr, not us. ### 2. OpenWhispr: the open-source original you are comparing against Worth keeping on your own shortlist. OpenWhispr does the core job well and does it in the open. It runs fully local transcription with whisper.cpp and Parakeet, bundles the binary so you do not need Python for basic use, stores your history in a local SQLite database, and collects no telemetry. It offers global-hotkey dictation with auto-paste and a dedicated AI-agent voice hotkey. Its free tier gives you unlimited local dictation with no weekly word cap, plus a cloud allowance of 2,000 words per week and 100-plus languages. Pro is $6.67/mo and adds unlimited cloud transcription, device sync, and API access. **Maintenance check:** v1.9.2 released 29 August 2026 with commits landing 1 September 2026, and the project carries close to 6,000 GitHub stars. This is one of the healthiest projects in the category. It also ships something worth naming: automatic call detection, starting meeting transcription itself when Zoom, Teams, or FaceTime opens. Yaps now records and transcribes meetings too, on macOS, Windows, and Linux, with on-device speaker labelling on macOS and Windows, but you start the recording yourself rather than have it auto-detect the call. **Best for:** developers and open-source purists on desktop who want auditable, self-hostable, unlimited local dictation, and live meeting notes. **Honest trade-off:** no Android app, and a power-user posture. No read-aloud or notes workspace. The iOS build is newer than the desktop app, so treat it as a companion rather than a proven equal. ### 3. Handy: the free, no-account open-source peer Handy is the other open-source project people compare directly against OpenWhispr, and on GitHub traction it is now the largest in the whole category, with over 30,000 stars. It is MIT licensed like OpenWhispr, but it goes further on the free question: there is no paid tier at all, and no cloud allowance to run out of. It ships for macOS on both Intel and Apple Silicon, x64 Windows, and x64 Linux, and runs Whisper (Small, Medium, Turbo, Large) with GPU acceleration plus Parakeet V3 on CPU with automatic language detection. Its stated position is that accessibility tooling belongs in everyone's hands rather than behind a paywall. **Best for:** anyone who liked OpenWhispr's philosophy and wants it with zero paid tier and a simpler surface. **Honest trade-off:** dictation only, no mobile app at all, no meeting notes, and no support desk. **Maintenance check:** commits landing 31 August 2026. ### 4. FluidVoice: the biggest Mac-native open-source option FluidVoice is a GPL-3.0 macOS dictation app that has picked up over 11,000 GitHub stars, which makes it the largest Mac-specific project of this shape. If you are on a Mac and want a second credible open-source option to weigh against VoiceInk, this is it. **Best for:** Mac users choosing between open-source dictation apps who want to compare more than one. **Honest trade-off:** macOS only, so it solves auditability but not platform reach, and a separate community Windows port exists that is a different project. **Maintenance check:** commits landing 1 September 2026. ### 5. MacWhisper: batch transcription of recorded audio MacWhisper is the veteran of Whisper-on-Mac and the pick when your job is turning stacks of recorded audio into text. It handles interviews, lectures, and podcasts well, with strong model selection and reasonable pricing. Where it is less of a fit is live, system-wide dictation and anything beyond Mac. It is macOS-only and does not carry a notes workspace or read-aloud. We cover it in depth in our [MacWhisper alternative](/blog/macwhisper-alternative) write-up. **Best for:** researchers and podcasters with a backlog of audio files to transcribe on a Mac. **Honest trade-off:** Mac-only, and built around file transcription more than everyday speak-to-type. ### 6. SuperWhisper: polished on-device dictation for Mac SuperWhisper is a clean, well-made on-device speech-to-text app for Mac. It runs Whisper models locally, gives you accurate transcription without a cloud dependency, and lets you pick different model sizes to trade speed against accuracy. Like OpenWhispr, it is a strong privacy peer. Where it differs from Yaps is scope and platform: it is Mac-only and focused on dictation, with no notes workspace, no read-aloud, and no Android. **Best for:** Mac users who want a focused, polished, on-device dictation tool and like tuning Whisper model sizes. **Honest trade-off:** Mac-only and single-purpose. No phone app, no notes, no read-aloud. ### 7. VoiceInk: the open-source Mac dictation frontend VoiceInk is another fully open-source, on-device dictation app for Mac. Like OpenWhispr, it is transparent and audit-friendly, and it is a good pick for people who want to read the code that processes their voice. It shares OpenWhispr's strengths and its limits: it is Mac-focused, dictation-first, and rougher on convenience features than a commercial app. Unlike OpenWhispr it is a paid one-time purchase ($25 to $49 depending on how many Macs) with a free trial download, it is GPL v3 rather than MIT, and it requires Apple Silicon and macOS 14.4 or later, so Intel Macs are out. If open-source-on-Mac is your priority, it belongs on the list. **Maintenance check:** v2.13 released 27 August 2026, commits landing 1 September 2026. **Best for:** developers and privacy maximalists on Mac who want open, inspectable dictation code. **Honest trade-off:** Mac-focused and dictation-only, with fewer convenience features than a polished commercial tool. ### 8. Wispr Flow: cloud dictation with strong reformatting Wispr Flow is the outlier here because it is cloud-based, not on-device. Its strength is AI reformatting: it rewrites raw speech into polished prose, and many people rate that quality highly. The trade-off is the one you were probably avoiding by looking at OpenWhispr in the first place. Every dictation leaves your device for the cloud, and it is subscription-only. If reformatting quality matters more to you than local processing, it is worth a look. If you wanted local for a reason, it is the wrong fit. **Best for:** users who prioritise cloud reformatting quality and do not mind sending audio off-device. **Honest trade-off:** cloud-based, so your audio leaves your machine, and it is subscription-only. How we ranked these We weighted platform breadth (does it work where you actually type, including your phone), on-device processing over cloud, scope beyond bare transcription, and how much setup it takes to get running. We build Yaps and we disclosed it. Where OpenWhispr or another tool genuinely wins for a specific need, we say so plainly. 3.0 ## Yaps versus OpenWhispr, side by side Both apps run transcription on your own device, and both avoid telemetry. So this table is not a "we are private and they are not" story, because that would not be true. It is a story about breadth, workspace, and setup. Every cell is accurate to what each app ships today. Scroll → | Feature | Yaps | OpenWhispr | | --- | --- | --- | | Platforms | Android, iPhone, iPad, Windows, macOS, Linux, Chrome ext | macOS, Windows, Linux | | Phone app today | Android, iPhone, iPad | None (iOS "coming soon") | | On-device transcription | Yes | Yes | | No telemetry | Yes | Yes | | Open source | No | Yes (MIT) | | System-wide dictation / keyboard | Hotkey + Android keyboard | Global hotkey (desktop) | | Audio-file transcription | Yes (Studio, to text/SRT) | Meeting recordings | | Live meeting transcription | Yes, desktop only (with diarization) | Yes, with auto-detect (with diarization) | | Read-aloud (TTS) | 18 desktop / 2 mobile voices | No | | Notes / vault with sync | Kanban, checklists, sync | Notes with search | | Setup | Download and go | Power-user (Node.js to build) | | Free tier shape | 7-day free trial | Unlimited local, 2K/wk cloud | | Price | Pro $15, Max $25/mo (7-day trial) | Free; Pro $6.67/mo | ![Diagram contrasting the open-source setup path of OpenWhispr with the install-and-go path of Yaps, both running on-device.](/images/blog/openwhispr-alternative-1.jpg) Read that table honestly and you can see the split. OpenWhispr wins on open source, on meeting transcription that auto-detects the call and starts itself, on unlimited local dictation with no weekly cap, and on price. Yaps wins on platform breadth (especially the Android phone app), on read-aloud, on the notes-and-sync workspace, and on installing without touching a terminal. Neither one is strictly better than the other, which is the whole point of running a comparison instead of a sales pitch. 4.0 ## Who should choose which The right answer depends on where you type and how much you want to tinker. ### Choose Yaps when you want to - Dictate privately on your phone today, on Android, not just on a desktop. - Keep your voice on your own device without cloning a repo or running a dev tool. - Get voice notes, audio-file transcription, and read-aloud in one app. - Sync notes between your phone and your computer. - Install and start dictating in a minute, with no setup. ### Stay with OpenWhispr when you - Require open-source, auditable, or self-hostable code (Yaps cannot match this). - Need meeting transcription that auto-detects Zoom, Teams, or FaceTime and starts itself. - Want truly unlimited local dictation with no weekly word cap. - Are a developer who wants an AI-agent hotkey, a public API, or an MCP server. A few edge cases route elsewhere. If your job is transcribing a backlog of recorded interviews on a Mac, MacWhisper is purpose-built for that. If you want a polished, focused, on-device dictation app for Mac and nothing more, SuperWhisper or VoiceInk fit. And if reformatting quality outranks local processing for you, Wispr Flow is the cloud option, with the trade-off that your audio leaves your device. For the broader landscape of local tools, our [best offline AI apps of 2026](/blog/best-offline-ai-apps-2026) roundup is a good next read. Private by design Everything Yaps hears stays on your device. Dictation, voice notes and read-aloud. Everything you say stays on your device. [Download Yaps](/download)iPhone · Android · Windows · Mac · Linux 5.0 ## Coming from OpenWhispr? Here is the switch If OpenWhispr has been your desktop dictation tool and you want to add a phone, notes, and read-aloud without giving up on-device processing, moving to Yaps takes about five minutes. You do not need to uninstall anything to try it. Step 01 ### Install Yaps and the on-device model one-time Download Yaps for your desktop, or grab it on Android. On first run it installs the on-device speech model. After that, dictation works with the internet switched off. Step 02 ### Set a hotkey that does not clash 30 sec On desktop you push the Yaps dictation hotkey: Fn on macOS, Ctrl and the Windows key on Windows, or a double-tap of Ctrl on Linux, all rebindable in settings. On Android you tap the dictation button on the Yaps keyboard. Keep OpenWhispr on its own hotkey so the two do not fight while you compare. Step 03 ### Try the parts OpenWhispr does not have a few days Import a recording into Studio and export a transcript or SRT. Capture a few voice notes into Kanban or a checklist. Highlight a draft and hear it read aloud before you send. Step 04 ### Pair your phone and decide 2 min Scan the QR to pair vault sync between phone and desktop. Run both apps for a week, then keep whichever fits. If you need open source, that is a fair reason to stay with OpenWhispr. 01 · Try Yaps ## A voice keyboard that keeps your voice on your phone. Install Yaps on Android for offline dictation, a familiar full-size keyboard, and no screen capture. Scan the QR on desktop, or tap the Play badge on mobile. [GET IT ON Google Play](https://play.google.com/store/apps/details?id=ai.yaps.app) or [yaps.ai/android](https://play.google.com/store/apps/details?id=ai.yaps.app) 6.0 ## Frequently Asked Questions ### Is OpenWhispr free? Yes, in a real sense. OpenWhispr is free to download, and its free tier gives you unlimited local dictation with no weekly word cap, plus 2,000 words per week of cloud transcription. The paid Pro tier at $6.67/mo mainly adds unlimited cloud transcription, device sync, and API access. If you only ever use local models, you can run it at no cost forever. ### Is OpenWhispr open source? Yes, OpenWhispr is fully open source under an MIT license, which permits personal and commercial use. Its code is on GitHub under the OpenWhispr organisation, and the repo states there is no data collection and no telemetry. This is a genuine advantage over closed-source tools, including Yaps, if auditable code matters to you. ### What platforms does OpenWhispr support? OpenWhispr runs on macOS (Apple Silicon and Intel), Windows, Linux, and iOS. Desktop builds are packaged as AppImage, .deb, .rpm, and .tar.gz. There is no Android app. Pages that still say "iOS coming soon" are out of date, but the Android gap is real and is the main reason people look at an alternative like Yaps. ### Does OpenWhispr work offline, or is it fully local? It is genuinely local. OpenWhispr runs transcription on your device using whisper.cpp and Parakeet, and its own materials state that audio never leaves your device. Transcription history is stored in a local SQLite database. Cloud processing exists but is optional and bring-your-own-key. On this point OpenWhispr and Yaps are peers, since both process core dictation on-device. ### Does OpenWhispr have a mobile app for iOS or Android? iOS yes, Android no. OpenWhispr's site now lists iOS alongside its macOS, Windows and Linux builds, so the "coming soon" line in older comparisons is stale. There is still no Android app at all. If you dictate on an Android phone, Yaps ships a full Android voice keyboard with on-device dictation, and that gap is the clearest single reason people search for an OpenWhispr alternative. ### Is OpenWhispr still maintained in 2026? Yes, OpenWhispr is one of the most actively maintained projects in this category. Version 1.9.2 was released on 29 August 2026, commits were landing on 1 September 2026, and the repository carries close to 6,000 stars. We checked this directly rather than taking a website's word for it, because several dictation projects in this space still have live marketing sites and no code activity behind them. ### Which OpenWhispr alternatives are open source and still shipping? Four are open source and all four were shipping code in the last week of August 2026: Handy (MIT, commits 31 August 2026, 30,000-plus stars), FluidVoice (GPL-3.0, commits 1 September 2026, 11,000-plus stars), VoiceInk (GPL v3, v2.13 on 27 August 2026), and OpenWhispr itself (MIT, v1.9.2 on 29 August 2026). Two smaller ones worth knowing are Amical (MIT, commits 26 August 2026) and VoiceTypr (AGPL-3.0, commits 27 August 2026). Yaps is not open source, and we say so plainly. ### OpenWhispr vs Handy, which should I pick? Pick Handy if you want the simplest free path, because it has no paid tier at all and no cloud allowance to run out of; pick OpenWhispr if you want meeting notes, an iOS build, device sync, and a cloud option through your own keys. Both are MIT licensed, both run Whisper and Parakeet locally, and both are actively maintained, so this is a scope decision rather than a privacy one. ### How do you set up local Whisper models in OpenWhispr? For most people, you install a prebuilt binary and the local models auto-download on your first transcription into a cache folder. That part is low-friction. Building from source, on the other hand, means cloning the GitHub repo and using Node.js 24 or newer, which is why OpenWhispr describes itself as a developer-friendly, power-user tool. If you would rather skip all of that, Yaps installs the on-device model on first run and needs no terminal. ### What is the difference between OpenWhispr's Whisper and Parakeet models? Both run on-device, and the difference is which engine transcribes your speech. OpenWhispr uses whisper.cpp, a local build of OpenAI's Whisper, and sherpa-onnx running NVIDIA's Parakeet. In practice you pick based on the speed and accuracy trade-off you prefer for your machine and language. Yaps handles on-device model selection for you and does not ask you to choose an engine. ### OpenWhispr vs Wispr Flow, which is better? They sit at opposite ends. OpenWhispr is open-source and on-device, so your audio stays on your machine, and it positions itself directly as the open-source Wispr Flow alternative. Wispr Flow is cloud-based with strong AI reformatting, so your audio leaves your device but the polished-prose output is a genuine strength. If privacy and local processing matter most, OpenWhispr wins; if reformatting quality is the priority, Wispr Flow makes its case. ### OpenWhispr vs SuperWhisper, which is more private? Both are on-device and private, so neither is clearly "more private" than the other for core dictation. The real differences are elsewhere: OpenWhispr is open source and cross-platform on desktop (macOS, Windows, Linux), while SuperWhisper is a polished, closed-source, Mac-only app with granular Whisper model selection. Choose on openness and platform, not on a privacy gap, because there is not a meaningful one. ### Does OpenWhispr collect data or have telemetry? No. OpenWhispr's repo states it has no data collection, no telemetry, and zero data retention, and its transcription runs locally with history in a local database. When it uses optional cloud processing with your own API key, it reports that cloud audio is never stored. This on-device, no-telemetry posture is one it shares with Yaps. ### What are the best OpenWhispr alternatives in 2026? Yaps is the strongest all-round pick for private voice on your phone and desktop with no setup, since it ships on Android alongside iPhone, iPad, Windows, macOS, Linux, and a Chrome extension, and it adds notes, audio-file transcription, and read-aloud. Handy is the best free open-source alternative, with no paid tier at all. FluidVoice and VoiceInk are the open-source Mac picks, MacWhisper is the one for batch transcription of recorded audio, SuperWhisper is the polished on-device Mac dictation app, and Wispr Flow is the cloud option if reformatting matters more than local processing. 7.0 ## Final thoughts OpenWhispr is a good, honest, open-source project, and if you live on the desktop, love open source, and want unlimited local dictation with live meeting notes, it earns its place. We are not going to pretend otherwise, and if auditable code is a hard requirement, it is the better tool for you. For most people who searched for an alternative, though, the reason is simple. You want private voice where you actually type, including an Android phone, and you would rather not run a developer tool to get it. That is where Yaps is the default: on-device dictation on Android, iPhone, iPad, Windows, macOS, and Linux, plus voice notes, [Studio audio-file transcription](/features/dictation), read-aloud, and vault sync, with no setup and no GitHub. If free and auditable is the harder requirement, install Handy instead; we would rather you had the right tool than ours. If you tend to hit dictation limits mid-thought, our take on [dictation with no time limit](/blog/dictation-app-no-time-limit) covers how Yaps handles longer sessions. Install it, try it against OpenWhispr for a week, and keep whichever one fits your life. The Yaps Journal is written with Yaps Put your voice to work. Messy speech in, clean ready-to-paste text out. Private, offline-first, and free to start. [Download Yaps](/download)iPhone · Android · Windows · Mac · Linux Keep reading ## Keep reading [COMPARISON · 13 MIN READ Local vs Cloud Translation for Confidential Documents](/blog/local-vs-cloud-translation) [COMPARISON · 13 MIN READ 9 Best AI Keyboards for iPhone in 2026 (On-Device Picks)](/blog/best-ai-keyboards-ios) **Yaps journal** · 2026 · Entry 99 [About](/about) [Brand](/brand) [Careers](/careers) [RSS](/blog/rss.xml) [Archive](/blog) [Sitemap](/sitemap.xml)