API v2 is live — now with YouTube Live streaming + WebSocket support

AI systems are not limited
by reasoning anymore.
They are limited by
what they can structurally understand.

Don't Transcribe it — Loom it. Loose threads become the lasting fabric in AI Context building. Structured audio understanding for AI systems — typed JSON knowledge objects, not raw transcript text. One call. Any language. Built for AI systems, RAG pipelines, and developer tools.

Get started free View on GitHub

signalloom-quickstart

$ pip install signalloom

✓ Successfully installed signalloom

$ curl -X POST https://api.signalloomai.com/v1/transcribe \
-H "Authorization: Bearer $SL_API_KEY" \
-F file=@recording.mp4

✓ "status": "success" — processed in 1.4s

Scroll to explore

What it does

Video and audio in.
Structured knowledge out.

Pass in any media file — Signal Loom returns clean, typed JSON that AI systems can actually reason about. No more wrestling with raw transcript text.

Input

Upload any video or audio file. We'll handle the rest.

.mp4 .mp3 .webm .m4a YouTube* TikTok GDrive Dropbox .wav .ogg .flac

*YouTube link ingestion is best-effort due to platform anti-bot protections. Direct file links, Google Drive, Dropbox, Box, and similar hosted files are currently the most reliable.

Output

Structured JSON with speakers, topics, entities, and timestamps — ready for any AI pipeline.

Example API response

Copy

Why Signal Loom

Built for how AI actually works

Other transcription APIs give you plain text. Signal Loom gives you structured data that slots directly into your RAG pipeline, agent memory, or knowledge graph.

Language-agnostic

Structured output works identically regardless of source language. Build global products without localization overhead.

Structured, typed JSON

Every response is a strongly-typed knowledge object. Speakers, topics, entities, sentiment — all pre-parsed and ready to use.

Fast, with live streaming

Transcription in seconds, not minutes. YouTube Live support with WebSocket streaming — same structured JSON as batch processing.

Affordable at scale

Generous free tier. No per-character pricing traps. Predictable pricing that scales with your usage, not your revenue.

Agent-ready output

Output is designed for AI consumption first. No post-processing needed. Just deserialize and use.

Developer-first DX

Clean REST API, official SDKs for Node/Python/Go, webhooks, OpenAPI spec, and docs you'll actually enjoy reading.

New — Live Streaming

Transcribe YouTube Live streams
as they happen.

Paste a YouTube live stream URL. Get structured JSON updates as the stream progresses. Same schema as batch transcription — no pipeline changes required.

How live streaming works

Start a live session

POST the YouTube live URL — we pipe the stream through yt-dlp → ffmpeg → Whisper in real time.

Receive partial transcripts

Structured JSON updates every 15 seconds. Speakers, timestamps, and text — all typed and ready for AI consumption.

Accumulated full transcript

Each update includes the running transcript so far. No need to stitch segments together yourself.

End session

Stop the session at any time. A final consolidated transcript is assembled from all partial updates.

# Start a YouTube live stream session
curl -X POST https://api.signalloomai.com/v1/live/start \
  -H "Authorization: Bearer $SL_API_KEY" \
  -d '{
    "url": "https://youtube.com/live/...",
    "chunk_duration": 15
  }'
# Returns: {"session_id": "ytlive_...", "state": "starting"}

Real-time WebSocket stream

WebSocket connection receives structured partial transcripts as the stream runs.

# WebSocket connection

$ wscat -c wss://api.signalloomai.com/v1/ws/stream?api_key=$SL_API_KEY

# Partial transcript — chunk 1

{

"type": "partial",

"session_id": "ytlive_abc123",

"chunk_index": 1,

"text": "Welcome everyone to today's...",

"accumulated_text": "Welcome everyone...",

"speakers": [{"id": "SPEAKER_00", "name": "Host"}],

"language": "en",

"duration_seconds": 14.8

}

# Partial transcript — chunk 2

{

"type": "partial",

"chunk_index": 2,

"accumulated_text": "Welcome everyone to today's stream. Today we have an important announcement...",

"text": "Today we have an important...",

"speakers": [{"id": "SPEAKER_00", "name": "Host"}]

}

YouTube Live supported

Streaming partial transcripts

Same structured JSON as batch

Speaker diarization included

WebSocket + REST both supported

No session time limits

Try live streaming free

Included in Starter plan and above. Free tier: 5 live sessions/month.

Who it's for

Built for builders

If you're building anything that needs to understand what people say — Signal Loom handles the hard part.

AI Agent Developers

Give your agent real audio/video understanding. Feed structured knowledge objects directly into your agent's context window instead of raw transcript text.

AI Agents

RAG Pipeline Engineers

Replace chunked text embeddings with richer signal — speakers, topics, entities, and timestamps that make retrieval dramatically more precise.

RAG Pipelines

Knowledge System Builders

Build organizational knowledge bases from meetings, calls, and video content. Structured output makes ingestion and querying far more reliable.

Knowledge Graphs

Developer Tool Makers

Add transcription to your IDE, CLI, or dev tool. Simple API, predictable pricing, and output that plays nice with any stack.

Developer Tools

By SignalloomAI™

One prompt. Every model.
Every cost. Before you run.

LoomLensAI™ shows you exactly what every AI model charges for your prompts — before you spend a single token. Built for developers and AI agents who want to make cost-informed decisions.

⚡

Instant Estimates

Paste any prompt. Get cost estimates across OpenAI, Anthropic, Google, DeepSeek, and xAI in milliseconds.

💰

Model Comparison

See exactly what Claude Opus costs vs Gemini Flash for your specific job. Make data-driven routing decisions.

🔗

API-First Design

Call POST /v1/estimate from any agent or pipeline. Token budgeting built into your workflow automatically.

Try LoomLensAI Free →

5 free estimates/day · Pro from $5/mo · Per-job $0.005

Pricing

Simple, transparent pricing

Start free, scale as you grow. No surprise bills, no character limits that punish natural speech.

Free

$0 /mo

100 min/mo in exchange for your email

✓ 100 min/month
✓ Up to 10 concurrent jobs
✓ REST API access
✓ JSON output
✓ Community support

Get started free

SignalloomAI™ Pro

🚀 40% OFF — First 59 days from launch (reg. $99/mo)

$59/mo

or $449/year

SAVE 37%

$449/yr

$37/mo equivalent

LOCKED IN 12 MO

✓ Full transcription platform
✓ ★ LoomLensAI™ Power Pro (1,000 est/mo)
✓ All future SignalloomAI tools included
✓ 5,000 min/month transcription
✓ Up to 30 concurrent jobs
✓ Webhooks + OpenAPI

Monthly: locks at $59/mo for 6 months, then $99/mo
Annual: locks at $449/yr for 12 months, then $699/yr

Claim Early Adopter Price →

Starter

$25 /mo

For indie devs and small projects

✓ 1,000 min/month
✓ Up to 15 concurrent jobs
✓ Streaming mode
✓ Speaker diarization
✓ Entity extraction
✓ Email support

Start building

Dev Pro

$349 /mo

For power users and solo builders

✓ 25,000 min/month
✓ Up to 50 concurrent jobs
✓ Everything in Pro
✓ Priority support

Get started

Enterprise

$1,595 /mo

Up to 5 seats · 100K mins/mo

✓ Up to 5 seats included
✓ Up to 100 concurrent jobs
✓ Everything in Dev Pro
✓ SSO / SAML
✓ Dedicated account manager
✓ Invoice billing / Net-30

Get started

Need a custom volume or on-prem deployment? Talk to us.

For AI Agents & Bots

Volume pricing for AI systems

Built for AI agents, sub-agents, and bots that process audio at scale. No seats, no per-seat pricing — just a monthly minute pool.

New — LoomLensAI™

See your AI costs before you run

One prompt. Every model. Every cost. Before you run. The token budget calculator built for AI developers and agents. Free 5/day. Pro $5/mo. Per-job $0.005.

Try LoomLensAI Free → See Pricing → Build with Us →

Built for AI agents, sub-agents, and bots that process audio at scale. No seats, no per-seat pricing — just a monthly minute pool.

Agent Free

$0 /mo

100 min/mo — no card required

✓ 100 included min/month
✓ $0.020/min overage
✓ Up to 10 req/min
✓ REST API access
✓ JSON output

Get API key

Agent Starter

$25 /mo

2,000 included min/mo

✓ 2,000 included min/month
✓ $0.015/min overage
✓ Up to 50 req/min
✓ Speaker diarization
✓ Entity extraction
✓ Email support

Start building

Agent Pro

$99 /mo

10,000 included min/mo

✓ 10,000 included min/month
✓ $0.010/min overage
✓ Up to 200 req/min
✓ Everything in Agent Starter
✓ Topic segmentation
✓ Sentiment analysis
✓ Priority support

Start building

Agent Dev Pro

$349 /mo

50,000 included min/mo

✓ 50,000 included min/month
✓ $0.007/min overage
✓ Up to 200 req/min
✓ Everything in Agent Pro
✓ Priority support

Get started

Agent Enterprise

$1,595 /mo

350,000 included min/mo

✓ 350,000 included min/month
✓ $0.005/min overage
✓ Up to 500 req/min
✓ Everything in Agent Dev Pro
✓ Custom SLA
✓ Dedicated account manager
✓ Invoice billing / Net-30

Bot-to-bot pricing? Ask about Loom Partners — up to 28% revenue share for AI platforms.

Getting started

Up and running in minutes

Three steps from zero to working transcription. No configuration, no infrastructure to manage.

Install the SDK

Pick your language. Our SDK handles authentication, retries, and response parsing automatically.

# Node.js
# npm install @signalloom/sdk  # Node SDK — coming soon
# Python
pip install signalloom
# Go
# go get github.com/signalloom/sdk-go  # Go SDK — coming soon

Get your API key

export SL_API_KEY=sl_live_••••••••••••••••
# Or pass it directly
SIGNALLOOM_API_KEY=sl_live_... your-script.sh

Call the API

One request. Your media file comes in, structured JSON comes out. That's it.

// Node.js — 5 lines of code
import { SignalLoom } from '@signalloom/sdk'
const client = new SignalLoom({
  apiKey: process.env.SL_API_KEY
})
const result = await client.transcribe({
  file: './meeting.mp4',
  speakers: true,
  entities: true
})
console.log(result.knowledge) // → structured JSON

AI systems are not limitedby reasoning anymore.They are limited bywhat they can structurally understand.

Video and audio in.Structured knowledge out.

Built for how AI actually works

Language-agnostic

Structured, typed JSON

Fast, with live streaming

Affordable at scale

Agent-ready output

Developer-first DX

Transcribe YouTube Live streamsas they happen.

How live streaming works

Real-time WebSocket stream

Built for builders

AI Agent Developers

RAG Pipeline Engineers

Knowledge System Builders

Developer Tool Makers

One prompt. Every model.Every cost. Before you run.

Simple, transparent pricing

Volume pricing for AI systems

See your AI costs before you run

Up and running in minutes

Install the SDK

Get your API key

Call the API

AI systems are not limited
by reasoning anymore.
They are limited by
what they can structurally understand.

Video and audio in.
Structured knowledge out.

Transcribe YouTube Live streams
as they happen.

One prompt. Every model.
Every cost. Before you run.