Python Livestream Toolkit

Python Livestream Toolkit provides two explicit YouTube playout modes:

• a minimal continuous terminal feed that captures one public market page at a fixed, high-quality 1080p30
• an optional narrated show that researches topics, generates scripts and TTS, and composes a branded studio page

The continuous terminal feed is the default container path. It navigates to STREAM_URL once and keeps Chromium and Xvfb alive while FFmpeg reconnects in process after transient output failures. The narrated runtime remains available through stream_url.py and its buffered producer-consumer pipeline.

What It Does

• Streams a public terminal page without exposing the host desktop
• Publishes 1920x1080 at 30 FPS with H.264 High, 10 Mbps CBR, a two-second GOP, and silent AAC audio
• Uses encrypted RTMPS for remote ingest and rejects plaintext remote RTMP endpoints
• Keeps stream credentials out of logs and out of Chromium/FFmpeg child environments
• Loads a reusable show profile from shows/*.toml
• Pulls source material from rss, webpage, json, or manual adapters
• Builds a local browser-based studio page with headline, source cards, ticker, and branding
• Buffers prepared audio segments ahead of playout
• Persists rolling memory per show to reduce repetition
• Routes script generation across xAI, Gemini, or OpenRouter
• Supports stable screenshot capture, macOS screen capture, and isolated virtual-display capture for containers
• Mixes continuous background music under narrated segments
• Supports optional music-only breaks between segments

Show Profiles

Each show profile defines:

• topic and editorial framing
• host voice and branding
• source adapters
• studio labels and optional iframe reference panel
• scene modes such as overlay, clean-feed, and transition
• segment rundown, for example headline, live feed, deep dive, and recap

Bundled profile:

• shows/hb_capital.toml

Capture Modes

Backend	Best for	Notes
playwright	stability	Default path, typically around 12 FPS
screen	local macOS experiments	Captures an entire display via avfoundation
virtual-screen	Docker or Linux VM	Captures only an isolated Xvfb-hosted Chromium session

If you do not want to stream your real desktop, use virtual-screen on Linux or in the provided container image.

Architecture

The runtime is split into small modules:

1. shows/ loads show configs, source adapters, and segment briefs
2. llm/ builds prompts and routes script generation
3. tts/ renders narration
4. broadcast/pipeline.py prepares queued segments and local studio pages
5. broadcast/streaming.py handles narrated FFmpeg playout
6. broadcast/memory.py records prior coverage context per show

The dedicated continuous path is split across stream_terminal.py and the small broadcast/terminal_stream*.py modules. See docs/terminal-stream.md for its operating contract.

More detail lives in docs/architecture.md.

Quick Start

Local install

git clone https://github.com/SebastianBoehler/python_livestream.git
cd python_livestream
pip install -r requirements.txt
playwright install

requirements.txt includes the full local-model stack. The default container intentionally uses the lighter requirements-stream.txt instead.

Environment

Copy .env.example to .env and set the values you need.

Common settings:

STREAM_URL=https://www.hb-capital.app/livestream
YOUTUBE_STREAM_KEY_FILE=/run/secrets/youtube_stream_key

For local development only, YOUTUBE_STREAM_KEY is accepted as a fallback. Production should mount the key as a readable, tightly permissioned file and set YOUTUBE_STREAM_KEY_FILE; if the file setting is present but unreadable or empty, startup fails instead of falling back to the environment.

Notes:

• SHOW_ID selects a profile in shows/
• SHOW_CONFIG_PATH can point to any custom TOML file
• STREAM_URL is always required by the continuous feed and by narrated profiles that reference it
• NEWS_SEGMENT_SECONDS is still supported as a global duration override
• hb_capital uses an HB Capital scene pack with overlay desk shots, clean-feed terminal views, and branded transitions

Provider credentials depend on which services you use:

• GEMINI_API_KEY
• XAI_API_KEY
• OPENROUTER_API_KEY
• ELEVENLABS_API_KEY
• HF_TOKEN

Run the continuous terminal feed

python stream_terminal.py

This entry point is Linux-only because it uses Xvfb and X11 capture. On macOS, use the container or a Linux VM.

For a receiver-free VM burn-in, set STREAM_OUTPUT_FILE=/tmp/terminal.flv, stop the stream after the soak window, and inspect it with ffprobe. This is mutually exclusive with STREAM_OUTPUT_URL.

Run the narrated show

python stream_url.py

Use a custom show file

SHOW_CONFIG_PATH=/absolute/path/to/my_show.toml python stream_url.py

Run portrait mode

STREAM_ORIENTATION=portrait python stream_url.py

Add a music-only break between segments

INTER_SEGMENT_MUSIC_SECONDS=20 python stream_url.py

INTER_SEGMENT_DELAY_SECONDS is still accepted as a compatibility alias.

Use isolated browser capture

STREAM_CAPTURE_BACKEND=virtual-screen STREAM_FPS=25 python stream_url.py

virtual-screen is Linux-only and is the recommended mode for long-running container or VM deployment.

Show Config Example

show_id = "my_show"
title = "Niche Desk"
tagline = "Automated coverage for one clear audience."
host_name = "Desk"
host_role = "Operator anchor"
description = "Explain what changed and why it matters."
base_prompt = "Cover the most relevant developments from the last 24 hours."
llm_system_instruction = "You are a sharp and concise anchor."
tts_voice = "Charon"

[branding]
primary_color = "#93f5d8"
accent_color = "#f6c35c"
background_start = "#04141a"
background_end = "#11172a"
card_background = "rgba(8, 18, 31, 0.78)"
text_color = "#f4f7fb"
muted_text_color = "#9bb4c8"

[studio]
label = "Live Desk"
strapline = "Source-driven coverage"
ticker_prefix = "Radar"
iframe_url = ""
layout_mode = "split"

[[sources]]
kind = "rss"
name = "Google News Topic"
url = "https://news.google.com/rss/search?q=my+topic+when:1d&hl=en-US&gl=US&ceid=US:en"
limit = 5

[[sources]]
kind = "manual"
name = "Editorial Guardrails"
text = "Stay concrete and explain implications."

[[segments]]
kind = "headline"
label = "Top Setup"
instructions = "Open with the biggest development."
duration_seconds = 180
scene_mode = "overlay"

Docker

The default Dockerfile is optimized for the continuous terminal feed on Linux hosts:

• base image: python:3.11-slim-bookworm
• isolated Chromium capture via Xvfb
• fixed 1080p30 libx264 encoding for stable YouTube quality
• only streaming/runtime dependencies installed
• a non-root runtime user with Chromium's sandbox left enabled

Build and run it:

docker build -t python-livestream .
docker run --env-file .env python-livestream

For persistent operation:

docker compose up -d

The included docker-compose.yml uses the dedicated virtual-screen path and restart: unless-stopped.

The narrated show remains an opt-in Compose profile:

docker compose --profile narrated up narrated

GPU image

If you want a heavier local-model image for GPU-backed TTS experiments, use Dockerfile.gpu:

docker build -f Dockerfile.gpu -t python-livestream-gpu .

Repository Layout

broadcast/   streaming, capture backends, memory, intermissions, studio pages
docs/        architecture notes
llm/         provider routing and prompt generation
shows/       show configs, source adapters, segment brief helpers
tts/         TTS backends and chunking
tests/       unit tests for core streaming behavior

Testing

Run the baseline checks before opening a PR:

python -m py_compile $(git ls-files '*.py')
python -m unittest discover -s tests -p 'test_*.py' -v

CI also builds the default Docker image to catch container regressions early.

Contributing

See CONTRIBUTING.md for workflow and expectations.

In short:

• keep changes modular
• avoid large files and duplicated logic
• add or update tests for behavioral changes
• document operator-facing config changes

Security

If you discover a security issue, see SECURITY.md.

License

This project is licensed under the MIT License. See LICENSE.