A Whisper Transcription API.
Running on Your Own Computer.
Let your apps, scripts and automations send audio, video or a YouTube link to ScribeForge and get the transcript back — privately, offline, and without paying per minute.
Your Own Transcription Server. On Your Computer.
Other apps and scripts on your computer send audio or video to ScribeForge and get the transcript back automatically. Apps built for OpenAI's speech-to-text work by changing one address — they transcribe on your machine instead of in the cloud, with no per-minute bill.
- Drop-in replacement for OpenAI's
/v1/audio/transcriptions - Upload a file, use one already on disk, or pass a YouTube / video link
- Plain text, SRT, VTT or detailed JSON with timestamps
- One click to start — and it restarts when you reopen ScribeForge
Included in every paid plan, including the lifetime license.
# Transcribe a recording into SRT subtitles
curl http://127.0.0.1:47800/v1/audio/transcriptions \
-H "Authorization: Bearer YOUR_KEY" \
-F "file=@interview.mp3" \
-F "response_format=srt"from openai import OpenAI
# Same SDK — just point it at ScribeForge
client = OpenAI(
base_url="http://127.0.0.1:47800/v1",
api_key="YOUR_KEY",
)
with open("meeting.mp4", "rb") as f:
t = client.audio.transcriptions.create(model="whisper-1", file=f)
print(t.text)import OpenAI from "openai";
import fs from "node:fs";
const client = new OpenAI({
baseURL: "http://127.0.0.1:47800/v1",
apiKey: "YOUR_KEY",
});
const t = await client.audio.transcriptions.create({
model: "whisper-1",
file: fs.createReadStream("lecture.wav"),
});
console.log(t.text);Private by Design
The server only accepts connections from your own computer, and every request needs your private API key. Show, copy or regenerate it at any time.
Queued, Not Overloaded
Requests wait in a queue and run one at a time, so your computer stays responsive. Long files run in the background and report progress.
Saved to Your Library
Everything transcribed through the API lands in your library like any other transcript. Search it and download any transcript through the API too.
Plain-Language Guide
A "Getting started" tab walks you through setup in 3 steps, with a glossary and fixes for common problems — no developer background needed.
"Try It Now" Buttons
Check your connection and transcribe a test file through the API before you connect another app, so you know everything works.
Copy-Ready Examples
PowerShell, cURL, Python, Node.js and n8n examples with your address and key already filled in. A Server tab shows recent jobs and a request log.
Find it in the ScribeForge sidebar.
One click. Copy your address and key.
Paste the address and key into any OpenAI-compatible tool, script or n8n workflow.
Put Transcription Into Any Workflow
n8n & Automations
New recording in a folder or inbox? Send it to ScribeForge, then route the transcript to Notion, Slack, email or a spreadsheet.
Swap Out Cloud Whisper
Tools that already call OpenAI's speech-to-text can point at ScribeForge instead — same request, no cloud upload, no bill.
Video Link Pipelines
Pass a YouTube or other video link and get subtitles back as SRT or VTT, ready for captions, clips or show notes.
Scripts & Bulk Jobs
Loop over an archive from PowerShell or Python. Long files run in the background and report progress as they go.
Search Your Library
Query past transcripts and download any of them from your own tools — your library becomes a private, local data source.
Confidential Work
Legal, medical and research teams can automate transcription without a single file leaving the computer.
Endpoints at a Glance
Every route except the health check needs Authorization: Bearer YOUR_KEY. Full examples are inside the app.
/v1/audio/transcriptionsOpenAI-compatible: upload a file, get json, text, srt, vtt or verbose_json back.
/v1/transcriptionsTranscribe an upload, a file on disk, or a YouTube / video link as a background job.
/v1/jobs/{id}Check a job's progress and fetch the result. DELETE cancels it.
/v1/jobsList recent and queued jobs.
/v1/transcriptsSearch your transcript library.
/v1/transcripts/{id}Download any saved transcript.
/v1/modelsList the transcription models you have downloaded.
/v1/healthCheck the server is running. No key needed.
Local API Questions
What is the ScribeForge Local API?
A small server built into the ScribeForge desktop app. Other apps, scripts and automation tools on your computer can send it audio, video or a video link and get the transcript back — transcribed on your machine, not in the cloud.
Does it work with apps built for OpenAI Whisper?
Yes. It accepts the same request as OpenAI's speech-to-text endpoint (/v1/audio/transcriptions). Point the app at ScribeForge's address, use your ScribeForge key, and it transcribes locally instead.
Is it safe to leave the server running?
It only accepts connections from your own computer (127.0.0.1), and every request needs your private API key. You can show, copy or regenerate the key at any time.
Which model does the API use?
By default, the model you picked in ScribeForge Settings (requests for OpenAI's "whisper-1" use it too). You can also name any model you have downloaded; GET /v1/models lists them.
Are there per-minute fees or limits?
No. Transcription runs on your computer, so there is no meter. Requests are queued and processed one at a time so your computer stays responsive.
Which plans include the Local API?
The Local API is included in every paid ScribeForge plan, including the lifetime license, at no extra cost.
Do I need to be a developer?
No. The "Getting started" tab explains everything in plain language and walks you through setup in 3 steps, and "Try it now" buttons test the connection for you.
Unlimited Local Transcription for Your Apps.
The Local API is included with every paid plan. Pay once, transcribe as much as your apps need, and keep every recording on your machine.