To send a pre-recorded fireplace loop to YouTube Live with FFmpeg, create an encoder-based stream in YouTube Studio, then send the file to its ingest URL with the stream key. A file must be paced as a live feed: FFmpeg does not automatically send every file at real-time speed.
You will start FFmpeg, inspect the incoming picture and sound in Live Control Room, and start or confirm the event there. YouTube’s archive guidance describes what may happen after a stream; it is not a guarantee that a long-running loop will be saved.
Create an encoder-based YouTube live stream
First confirm that live streaming is enabled for your channel. YouTube’s current general guidance says a channel must be verified, have no live-streaming restrictions in the preceding 90 days, and be operated by someone at least 16 years old. If this is your first time enabling live streaming, activation can take up to 24 hours. Check the current YouTube live-streaming eligibility and setup guidance before choosing a broadcast time; do not assume an unenabled channel can go live immediately.
In YouTube Studio, choose the encoder workflow in Live Control Room rather than a webcam workflow. Create or schedule the live stream, add its title and visibility settings, and select the intended audience and other event details. You can schedule an event to establish a watch page in advance, or prepare an immediate event; either way, the encoder feed and the event’s start state are separate steps. Follow the on-screen state in Live Control Room when you are ready to publish.
For a fireplace station, check the full file before you build the stream. Confirm that it shows the intended scene throughout, that any transition between the end and beginning is acceptable, and that it has the audio you intend to broadcast. A fireplace image with crackling audio is still a video-and-audio programme for this purpose; a silent file is different from a file whose audio track is present but muted or defective. Keep a local copy of the original and note its resolution, frame rate, video codec, audio codec and duration.
A single loop can be suitable for a simple ambience channel, but it is not the same as a live camera or a varied programme. Think about whether repeated imagery and sound serve your viewers and whether you have the rights to use both. YouTube’s policies apply to live content as well as uploads. If you are building a devotional or regional channel around a continuous programme, this guide to setting up a 24/7 Telugu bhakti channel covers some of the broader channel decisions beyond the encoder itself.
Copy the stream URL and protect the stream key
In the event’s Live Control Room, locate the encoder settings. YouTube supplies a stream URL and a stream key. In FFmpeg, the URL is the ingest destination and the key is the credential used to associate and accept your feed. Treat the key like a password: do not put it in a public post, screenshot, shared document or command example that other people can access.
Copy the values exactly as shown. YouTube may show an ordinary RTMP address by default; if you want an encrypted connection, retrieve the RTMPS URL in the settings and use that exact address. YouTube recommends RTMPS, but the FFmpeg build you use must support the secure URL and connection. The official YouTube encoder settings guidance describes the ingest choices and encoding requirements. Do not replace a provided URL with one copied from an old tutorial, since event settings or available ingest options can change.
Keep the key out of any text you plan to publish. In a command typed directly into a terminal, remember that shell history may retain it. On a shared machine, avoid leaving credentials in a file or script that others can read, and use the operating system’s normal access controls. If you accidentally expose the key, reset or regenerate it in YouTube Studio and update the encoder destination before the next attempt.
YouTube’s control room is also where you will distinguish a feed arriving from an event that is actually live. Starting FFmpeg sends the encoder signal; it does not necessarily start the audience-facing event by itself. The preview and event controls give you a chance to catch a black picture, missing sound or wrong stream before viewers encounter it.
Prepare the fireplace loop as a real-time source
The key difference between playing a file locally and sending it to a live platform is timing. A normal file read can proceed as quickly as the computer and storage allow, which may push the whole programme through far faster than its intended duration. For a live broadcast, pace input to real time. FFmpeg documents -re as an input option for reading at the native frame rate, and its main documentation explains that option placement matters.
Check your installed FFmpeg version and documentation before building a loop command. In particular, verify the loop option and its value for that release rather than copying a flag from an unrelated example. The research basis for this guide does not claim a tested command or a universal loop setting. A reliable workflow is to test a finite section or one play-through first, then confirm the exact loop behaviour you intend to use. For a true continuous channel, plan what should happen at the file boundary and what you will do if FFmpeg exits.
Inspect the source with a media-information tool or FFmpeg’s output, and decide whether to pass through compatible streams or encode a new output. Stream-copying avoids encoding work, but only makes sense if the source’s formats and stream parameters meet YouTube’s ingest requirements. Re-encoding gives you control over codec, bitrate, frame rate and keyframes, while using more of the computer’s resources and creating another point to configure incorrectly. If you are uncertain, use a short test with the intended settings and inspect what YouTube receives.
For example, a 1080p file should not be treated as interchangeable with a low-resolution source just because the destination is the same. Choose output resolution and frame rate based on the file and the viewing purpose, then consult YouTube’s current bitrate table for that combination. Do not increase resolution merely to select a higher bitrate if the source contains no additional detail. Also check whether the audio track is present and set an appropriate audio encoder if you re-encode.
A local FFmpeg workflow means the computer running the process must remain on and connected for the broadcast. A power interruption, operating-system restart or network loss can stop the feed. If you are weighing that against a small always-on computer, this Raspberry Pi guide for pre-recorded YouTube streams may help frame the hardware choice; neither approach removes the need to check the actual file and connection.
Build the FFmpeg output using current YouTube guidance
Use YouTube’s current recommendations for the outgoing stream, not assumptions based only on the source file. Its RTMP/RTMPS guidance lists H.264 video, AAC or MP3 audio, constant bitrate (CBR), up to 60 frames per second, and a keyframe interval of two seconds recommended and no more than four seconds. The appropriate bitrate depends on the chosen resolution and frame rate, so use the current YouTube settings table rather than a number remembered from an older guide.
The command has four conceptual parts: input options, the input file, output encoding and pacing choices, then the output destination. FFmpeg applies options according to their position. Input options such as real-time pacing belong before the corresponding -i; output options belong after the input and before the destination. For the RTMP family, FFmpeg’s output format is commonly FLV. Its formats and protocol documentation is the reference for the muxer and its protocol documentation covers supported transports.
This is a structural template, not a tested command. Replace the file path and destination placeholders, and add encoding options that match YouTube’s current requirements and your selected source settings:
ffmpeg -re -i "/path/to/fireplace.mp4" [video/audio options] -f flv "<YouTube RTMP(S) ingest URL>/<stream key>"
The placeholder must never be a real key in material you share. With RTMPS, paste the exact secure ingest URL supplied by Live Control Room and check that your FFmpeg build can connect to it. With RTMP, the feed is not encrypted in the same way, so use YouTube’s recommended secure option when your setup supports it.
Do not paste the template unchanged and assume it encodes correctly. Square-bracketed text is explanatory, not valid FFmpeg syntax. Select a video encoder and its rate-control, frame-rate and keyframe options if re-encoding; select audio handling if the file contains audio. If you choose stream-copying, check that the existing tracks satisfy YouTube’s ingest requirements and that your FFmpeg output container can carry them. A successful local command is not proof that the incoming stream is healthy.
Bandwidth matters even with a prerecorded source. The outgoing bitrate is sustained while you broadcast, so a connection that only occasionally reaches the target is not a sound basis for an unattended channel. YouTube’s streaming tips recommend upload capacity around 20% above the total stream bitrate. Leave that headroom for normal variation, and test using the same connection and settings you expect to use for the full broadcast. If the source or connection proves difficult, the YouTube software options overview can help you compare workflows without treating one encoder as right for every channel.
Start FFmpeg and inspect the Live Control Room preview
Start with a controlled test rather than beginning a long unattended loop. Run FFmpeg with the exact file, output settings and destination you plan to use. Watch the terminal for errors and confirm that it continues reporting output instead of exiting immediately. Keep the process visible and note where its logs go, so an operator can identify whether a failure comes from file access, encoding or the network.
Then return to Live Control Room and wait for the incoming preview and stream health indicators. Confirm the fireplace image is moving as expected, the framing is correct, and audio is present at a sensible level if the file has sound. A frozen still, black frame, silent soundtrack or clipped audio is easier to fix before starting the event. YouTube recommends testing with similar audio and motion to the planned broadcast, rather than assuming a still-image test is representative.
A preview can take time to appear, and a preview is not the same as the public event. Do not start the event until you have checked what YouTube is receiving and are satisfied with the event details. If the preview does not appear, check the stream key and URL, the selected protocol, local firewall or network conditions, and FFmpeg’s error output. Change one thing at a time, then reconnect and observe the result.
Once the picture, sound and health state are acceptable, use the control room to start or confirm the live event according to its displayed state. Keep the terminal and control room accessible while you watch the first part of the broadcast. This is a handoff between encoder and platform, not a reason to close the computer or assume the programme will continue unattended.
Confirm the live event and test the broadcast
After starting, verify the audience-facing watch page if appropriate for the event’s visibility. Check that playback begins, that sound is in sync with the image and that the loop boundary does not create an unwanted pause or abrupt change. A control-room preview confirms what is reaching YouTube’s ingest, but checking the playback path gives you another way to spot issues that are not obvious in the local process.
YouTube’s streaming tips advise testing in advance and monitoring stream health. Treat monitoring as an operating task: decide who can see the alerts, who can restart the process, and what the response should be if the feed stops. For a small business or devotional channel, a simple written checklist with the file path, event name, start procedure and recovery steps is more useful overnight than relying on one person’s memory.
A local encoder depends on the host computer staying available. Disable settings that would put the machine to sleep during the broadcast, make sure the power supply is dependable, and avoid installing updates or restarting in the middle of the event. If the connection drops, FFmpeg may not recover in the way you expect; verify the behaviour on your version and have a way to check the feed again in Live Control Room. YouTube’s stream-health indicators can help diagnose incoming quality problems; this guide to yellow or red stream health in Studio is useful when the feed connects but reports trouble.
For a channel intended to run overnight, consider what an interruption looks like to the viewer and how you will learn about it. A router renewal, power cut or application exit can take the broadcast offline even if the video file is fine. If your home connection is part of the setup, review these steps to prevent an interruption during a router DHCP renewal. No test proves future continuity, but it can expose avoidable problems before you depend on the stream.
Understand archive guidance and its limits
YouTube says streams under 12 hours are automatically archived. Read that as platform guidance for streams below the stated duration, not as a promise that every broadcast will be saved, that a continuous loop will remain available as a recording, or that an interrupted stream will produce one complete archive. Archive availability is separate from keeping the live feed online.
If you need a copy of the programme, retain the source file and consider a separate recording plan. Confirm the current official guidance for your event and channel, particularly before planning a broadcast that approaches or exceeds the stated duration. Do not build a schedule that depends on an assumed archive as your only copy; the stream and the original media serve different purposes.
You also need to end the event deliberately. Stop the encoder feed when the programme is finished, then use Live Control Room to end the stream and confirm its status. For a loop intended to continue, do not confuse restarting the file with restarting the YouTube event. The operational plan should say what you intend to do after a disconnection and who is responsible for checking whether the event is still live.
Finally, confirm you have permission for every part of the broadcast, including the video and any soundtrack. The fact that a fireplace clip is quiet, old, or freely accessible does not establish rights to stream it. Review YouTube’s current Community Guidelines and Terms of Service for the channel, and do not assume that following encoder settings guarantees policy approval or monetisation.
Before committing, compare the operating options on the pricing page. When the file and channel are ready, start free — 24-hour trial, no card.
FAQ
Does FFmpeg play the file at real-time speed by default?
No. A file can be read as fast as the computer can process it unless you pace the input for real time. FFmpeg documents -re for real-time input reading; check its placement and test the resulting feed before going live.
Should I choose RTMP or RTMPS?
YouTube recommends RTMPS for an encrypted connection, provided your FFmpeg build supports the exact secure URL supplied in Live Control Room. Use the URL and key shown for your event rather than a copied destination from another guide.
Will YouTube definitely archive my fireplace stream?
No. YouTube’s guidance says streams under 12 hours are automatically archived, but that is not a guarantee for every broadcast or a promise about continuous loops and interruptions. Keep the original file and check current official guidance if you need an archive.
Can I leave the computer unattended once the preview looks right?
A good preview confirms that the feed is arriving at that moment; it does not ensure the computer, power or network will stay available. Keep the machine from sleeping, monitor stream health and have a recovery plan for a stopped process or disconnected feed.