Speakr: self-hosted transcription and note-taking
Don’t want to install Speakr? Drop the file here.
The first 20 minutes of any recording are transcribed free: a recording up to 20 minutes transcribes in full, a longer one is previewed, and a pack from $6 (500 minutes) transcribes the rest and opens every export, and credits never expire.
A self-hosted web app that records or ingests audio, sends it to the speech engine you configure, and turns transcripts into summaries, events and searchable notes for a team.
Get Speakr Official site & pricing
Speakr is a self-hosted library for recordings: it hands audio to the engine you configure (WhisperX, OpenAI, Voxtral, AssemblyAI and more), then summarises, tags and searches.
It is a platform to operate, not a model: Docker, an ASR container or API key, and an LLM key. For one person with a pile of recordings, a hosted transcript plus summary is the shorter path.
What it is
Speakr is a web application you host yourself, licensed under AGPL-3.0 or a separate commercial licence from its maintainers. It records from the microphone or system audio, or takes dropped files and a watched folder, then sends the audio to a transcription connector you choose: self-hosted WhisperX (which enables speaker labels and voice profiles), OpenAI, Mistral Voxtral, AssemblyAI, FunASR and others. A separate text model writes summaries, extracts events and powers chat across the library. It adds folders, tags with their own prompts, retention policies, OIDC single sign-on, a REST API and signed webhooks. The latest release is v0.10.13-alpha (4 October 2026).
Watch out for: Needs a speech engine (a self-hosted WhisperX container or an API key) and a text model for summaries. The full image is about 4.4 GB; releases are still tagged alpha.
The transcript and summary, without the deployment
Speakr gives a team its own platform. If you need the transcript and a summary of recordings you already have, WhipScribe does both in one job.
- Drop the audio or video file into the box at the top of this page, or paste a link (YouTube, a podcast or RSS feed, Google Drive, Twitch or Kick). Nothing to install and no GPU needed.
- The language is detected automatically and the transcript stays in the language that was spoken (WhipScribe does not translate).
- Every job comes back with the transcript plus a summary, chapters with timestamps and key quotes; speakers are named there.
- Read the summary free. A pack from $6 (500 minutes) opens the full transcript and the downloads: TXT, DOCX, SRT, VTT and JSON. Packs are one-time purchases and credits never expire.
- Files can be up to 10 hours or 5 GB per file.
Related: Bulk transcription · Transcribe large files · SRT generator · WhisperX · Pricing
Install / use
Go to Speakr murtaza-nasir.github.ioKey specs
What people actually do with Speakr-style transcription
The tool is the means. These are the jobs — each one priced at published rates, each one wired up on its own page.
Features
| Speaker diarization | Yes |
| Word-level timestamps | Not stated |
| Streaming / real-time | Not stated |
| Languages supported | — |
| HIPAA eligible | No |
Links
- github.com/murtaza-nasir/speakr ↗AGPL-3.0 · Python · v0.10.13-alpha on 2026-10-04
- Speakr documentation ↗quick start, connectors, companion apps
Speakr vs Whipscribe
| Feature | Speakr | Whipscribe |
|---|---|---|
| Category | Open source | Transcription APIs |
| Pricing | free (AGPL-3.0); you pay any speech or LLM API you connect | $6–$24 one-time packs (credits never expire) · first 20 minutes free |
| Speaker identification | Yes | Named in the summary, key quotes and chapters |
| Word timestamps | No | Yes |
| Streaming | No | No |
| Languages | — | 99 |
| Platforms | Docker, Web | Web, API, MCP |
Alternatives to Speakr
Frequently asked about Speakr
Does Speakr transcribe on its own?
No. Speakr orchestrates: you connect a transcription engine (a self-hosted WhisperX container, or an API such as OpenAI, Mistral Voxtral, AssemblyAI or FunASR) and a text model for summaries, titles and chat.
Does Speakr identify speakers?
Yes, with a diarizing backend: self-hosted WhisperX (which also enables voice profiles across recordings) or OpenAI's diarizing model, among others listed in its README.
How do I install Speakr?
With Docker: download its example docker-compose file and env template, add your transcription and text-model settings, and run docker compose up. A lite image (about 725 MB instead of 4.4 GB) skips PyTorch at the cost of semantic search.
Is Speakr free?
The software is free under AGPL-3.0; the maintainers also offer a commercial licence for anyone who cannot meet the AGPL's terms. Running it costs your hardware plus whatever the connected speech and LLM APIs charge.
Speakr or WhipScribe?
Speakr is the choice for a team that wants its own recording library with SSO and retention rules. WhipScribe is a hosted service: transcript, summary, chapters and SRT from a file or link, with nothing to deploy.
What is Speakr?
Speakr is a self-hosted library for recordings: it hands audio to the engine you configure (WhisperX, OpenAI, Voxtral, AssemblyAI and more), then summarises, tags and searches.
Does Speakr work on Mac, Windows and Linux?
Speakr runs on Docker and Web. It is not listed for other platforms; check the vendor for anything beyond Docker and Web.
Is Speakr HIPAA compliant?
Speakr does not claim HIPAA compliance. For PHI, run it on infrastructure you control or choose a service that will sign a BAA.
What are the limitations of Speakr?
Needs a speech engine (a self-hosted WhisperX container or an API key) and a text model for summaries. The full image is about 4.4 GB; releases are still tagged alpha.
Who is Speakr best for?
Teams that want a private, multi-user recording library with SSO, retention rules and summaries, and have someone to run Docker.
What should I know before choosing Speakr?
It is a platform to operate, not a model: Docker, an ASR container or API key, and an LLM key. For one person with a pile of recordings, a hosted transcript plus summary is the shorter path.
Does Speakr run offline?
Speakr runs on your own machine — it is open source, so nothing leaves the computer unless you configure it to.
Where is the Speakr source code?
On GitHub at github.com/murtaza-nasir/speakr. It is maintained by murtaza-nasir.
What kind of tool is Speakr?
In this directory Speakr is filed under self hosted, note taking as a open-source tool.
Can I automate Speakr-style transcription without running it myself?
Yes. If what you want is transcripts rather than the tool itself, Whipscribe does the same job as a hosted API: submit a file or URL, poll a job id, pull the result as txt, json, srt, vtt or docx. $0.012 a minute — about $0.72 an audio hour — bought as credits that never expire. The automation recipes on this site show it wired to Drive folders, Zoom recordings, S3 buckets and no-code platforms.
Whipscribe is a hosted transcription service. If you want transcripts without running infrastructure, paste a URL or drop a file in the form below — you'll have a transcript in seconds.
Transcribe your recordings
Running Speakr yourself? Send the same file here and compare the transcript. Transcribe any language, any recording · Transcription services from $0.29 an hour · Transcribe from any meeting or recording platform · How to transcribe, step by step
Explore
All transcription tools · By language · By platform · By file format · How-to guides · Transcription services · Transcription API · Pricing
Long recording to get through?
One file up to 10 hours, as an upload or a link: see MP3 to text for long recordings, how to transcribe a podcast and the SRT file generator.
Looking at paid options?
Before you choose a transcription tool or service: see how the best transcription services compare, cheap transcription services and the real cost per hour and how our transcription services work.