For a typical stereo internet-radio station sent to YouTube Live over RTMP or RTMPS, AAC is the clearest default. YouTube also lists MP3 as supported for those protocols; if you use HLS instead, choose from its separate list of AAC, AC3 and EAC3.
For either protocol, YouTube’s recommended stereo settings are 44.1 kHz sample rate and 128 kbps audio bitrate. These are ingest recommendations, not a promise of better sound or delivery: first confirm which protocol your encoder uses, then match the codec and settings to it.
Choose the ingest protocol first
The format question is not only about the audio file or codec. YouTube receives a live feed through an ingest protocol, and the protocol determines which audio codecs it lists as supported. A setting that is appropriate for RTMP/RTMPS should not be carried across to HLS without checking the HLS guidance.
RTMP and its secure extension, RTMPS, are the usual starting point for a conventional encoder workflow. YouTube recommends RTMPS in its general live encoder settings guidance. If your encoder already sends RTMP or RTMPS, you can usually focus on selecting a listed codec and setting the stereo sample rate and bitrate.
HLS is another ingest option, but it has distinct setup requirements. YouTube’s HLS setup instructions describe playlists and media segments, including a documented segment duration of 1–4 seconds. Because HLS sends segments rather than a continuous stream, YouTube notes that it has higher latency. That can matter for live interaction; it does not make HLS inherently wrong for every station.
| Ingest choice | YouTube-listed audio codecs | Recommended stereo settings | Practical consideration |
|---|---|---|---|
| RTMP/RTMPS | AAC or MP3 | 44.1 kHz, 128 kbps | A straightforward fit for many encoder workflows; YouTube recommends RTMPS. |
| HLS | AAC, AC3 or EAC3 | 44.1 kHz, 128 kbps | Requires HLS playlist and segment configuration; expect higher latency than a continuous stream. |
The table is a decision aid, not a claim that every encoder offers every combination. Check your encoder’s own protocol and codec options. If you are assembling a devotional loop, for example, a simple RTMPS setup with AAC stereo is a practical default; the guide to making a 24/7 church prayer stream can help with the broader channel workflow.
RTMP/RTMPS audio codec options
For RTMP and RTMPS, YouTube lists AAC and MP3 as supported audio codecs. For a new or adjustable stereo setup, choose AAC unless a concrete compatibility reason points elsewhere. This is a clear default, not a claim that AAC guarantees better sound, uninterrupted delivery or a particular viewer experience.
If your station software is already configured to encode MP3, you do not need to switch simply because AAC is the default recommendation here. YouTube lists MP3 as supported for RTMP/RTMPS too. Confirm that the encoder is actually sending the codec you selected, rather than assuming the source audio’s file extension determines the transmitted live codec.
A radio automation system may play MP3 files while producing an AAC live output. Those are separate stages: the source file format describes stored audio, while the output codec describes what the encoder sends to YouTube. Your goal is to configure the live output, not to convert every archived track without a reason.
Keep the stereo channel layout in view. A typical music, bhajan, ambience or spoken-radio station uses stereo audio; mono narration may be intentional, but do not treat a stereo recommendation as a requirement to manufacture stereo from a mono source. Check both channel layout and codec in the encoder’s output settings.
RTMPS is the secure extension YouTube recommends in its general encoder workflow. It is a protocol choice, not an audio codec. Selecting RTMPS does not by itself change the codec list: use AAC or MP3 as YouTube lists for RTMP/RTMPS, then apply the stereo settings separately.
HLS audio codec options
For HLS ingestion, YouTube lists AAC, AC3 and EAC3. This is a different set from the RTMP/RTMPS list, so do not infer that MP3 is an HLS option from its support for the other protocol. Choose a codec that both your encoder and the HLS workflow support, and verify that the emitted stream matches the selection.
AAC is available in both lists, which can make it a convenient choice if your encoder and operating workflow already use it. AC3 and EAC3 are listed HLS options as well; their presence is not a reason to select them for an ordinary stereo radio station if you have no specific need or compatible workflow. The useful question is not which name sounds more advanced, but which supported option your encoder can reliably produce for the protocol you chose.
HLS involves more than codec selection. You need to provide the expected playlists and segments, and YouTube documents segment durations of 1–4 seconds. Its segmented delivery has higher latency than a continuous stream. Consider that trade-off if listeners are following a live news bulletin or taking part in a call-in format; for a non-interactive music station, the delay may be less important than compatibility with your chosen workflow.
Do not copy RTMP/RTMPS settings into an HLS encoder by habit. Revisit the protocol-specific HLS setup page when configuring playlists, segments and codec. The broader OBS setup guide for a 24/7 YouTube stream may help with an encoder workflow, but use the official HLS requirements for protocol details.
YouTube’s stereo sample-rate and bitrate guidance
YouTube recommends 44.1 kHz sample rate and 128 kbps audio bitrate for stereo over both RTMP/RTMPS and HLS. If you need a concise starting point for a typical stereo internet-radio stream, set those values in the encoder after selecting the protocol and a codec listed for it. They are recommended advanced settings, not a guarantee that every part of your production chain will sound a certain way.
Sample rate and bitrate describe different things. The sample rate is the number of audio samples represented per second; bitrate is the data rate allocated to the encoded audio. Setting one correctly does not compensate for a mismatch in the other, and neither repairs a noisy source, clipping, silence or a poor mix.
Start with the source. If you feed a low-quality or distorted recording into an encoder, selecting the recommended output settings cannot recover detail that is not present. Likewise, raising the bitrate beyond YouTube’s recommendation is not automatically useful. Keep the settings aligned with the published guidance and spend your checks on the whole path: source playback, channel layout, encoder output and the received preview.
The figures here concern audio in a live encoder workflow. YouTube also receives video for a typical live stream and transcodes live streams into multiple playback formats for viewers. That means viewers do not necessarily receive the exact ingest stream unchanged. Choose a valid ingest configuration, then judge the actual preview and playback rather than treating the encoder’s output as the only version that matters.
This article is about radio-style stereo. If you are deliberately configuring 5.1 surround, do not reuse the stereo numbers: YouTube’s general guidance gives 48 kHz and 384 kbps for 5.1. Codec listings also remain protocol-specific: YouTube lists AAC only for RTMP/RTMPS 5.1 audio, while its HLS guidance lists AAC, AC3 and EAC3. Most internet-radio operators can ignore this exception unless they are producing a genuine surround feed.
Match encoder settings to the protocol
Make the protocol selection explicit before touching audio settings. In the encoder, identify whether the destination is RTMP/RTMPS or HLS. Then select a codec from the corresponding YouTube list, set stereo to 44.1 kHz and 128 kbps, and confirm that the encoder is not overriding those choices through a separate profile or preset.
Names vary between encoders, so look for terms such as audio encoder, output codec, sample rate, channels and bitrate. A preset called “YouTube” is not enough evidence by itself: inspect its actual values. If it offers a choice between AAC and MP3 for an RTMP/RTMPS feed, either is listed as supported; for a fresh stereo configuration, AAC is the simple default.
A useful preflight note for a small station can fit on one line: protocol, codec, channels, sample rate and bitrate. For example: “RTMPS; AAC; stereo; 44.1 kHz; 128 kbps.” This does not replace checking the live output, but it gives anyone taking a shift a clear reference and makes it easier to spot an accidental profile change after a software update.
Keep audio and video troubleshooting distinct. If the video freezes while audio continues, changing an audio codec is unlikely to be the relevant first step. If listeners report a problem that is absent in your control room, compare the received stream with your local encoder output; the guide to freezing for viewers in India but not in the control room covers that separate diagnostic situation.
If your stream is a loop built from a media file, test the entire loop rather than only a short track. Transitions can expose gaps, abrupt volume changes or a source that stops unexpectedly. For a stream that ends when its file reaches the end, check the OBS loop-source settings guide; codec selection will not make a source repeat.
Check the resulting stream
Before making the broadcast public, test with audio and movement similar to the programme you intend to run. YouTube recommends testing before going live and monitoring stream health and messages during the event. A brief test using only a quiet intro does not tell you whether a music bed, spoken segment or transition will behave as expected.
Listen to the live preview, not only the source file on the computer. Check that both stereo channels are present where expected, that dialogue is intelligible over music, and that there are no obvious dropouts or clipping. Watch the stream-health indicators and any encoder or YouTube messages. These checks can reveal a mismatch or source problem; they do not certify the stream for every listener or network condition.
For an always-on channel, add the audio checks to the handover routine. Record the selected protocol and values, and ask the next operator to confirm that the preview is receiving audio after a restart or configuration change. If the stream drops, inspect the messages and identify whether the issue is source playback, encoder output, network connection or ingest configuration before changing several settings at once.
Continuous operation also means deciding who or what notices a failure when the person who started the stream has gone to bed. For a file-based station, StreamNeo removes the need to leave your own computer running by taking an uploaded video and running it as a YouTube live stream, with monitoring and automatic restarts if it drops. It is YouTube-only, so it suits that specific workflow rather than a station that needs a different platform or a live, on-site mixer.
When you change audio settings, test again before relying on them overnight. Save the known-good configuration, keep a short representative test file available, and check the live preview and health messages after the change. That is a more useful safeguard than assuming a codec name alone settles the quality or reliability question.
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
What audio format should I use for internet radio on YouTube?
For a typical stereo station sent over RTMP or RTMPS, AAC is the clearest default, with YouTube’s recommended settings of 44.1 kHz and 128 kbps. If you use HLS, choose from its listed options—AAC, AC3 or EAC3—and apply the same stereo recommendations.
Is AAC or MP3 better for YouTube Live?
YouTube lists both AAC and MP3 as supported audio codecs for RTMP/RTMPS, so MP3 is not unsupported. If you are choosing a fresh setup for stereo, AAC is a straightforward default; retain MP3 if your existing encoder workflow uses it and has been checked.
Can I use MP3 with HLS?
YouTube’s HLS audio list is AAC, AC3 and EAC3, not MP3. Check the current HLS setup guidance and your encoder’s compatibility rather than carrying the RTMP/RTMPS list over to HLS.
Does the ingest codec determine what every viewer hears?
Not necessarily. YouTube says it transcodes live streams into multiple playback formats, so the viewer’s playback version need not be the exact ingest stream unchanged. Test the received preview and monitor stream health, while remembering those checks do not guarantee a particular sound or delivery outcome.