We should have a fix out for this very soon, thanks again for reporting it
We should have a fix out for this very soon, thanks again for reporting it
RHi! Can someone shed some light on how simulcasting is billed? I’m planning to Use Cloudflare stream to send one SRT stream to CF and feed multiple platforms off it using the simulcast (restream) option. The documentation only seems to mention payments for serving and storing video but I won’t be doing that on CF Stream.
PGlad you're watching this. I would be grateful if you let me know when you fix it. Thank you.
Btw, I ran the Apple HLS analyzer and it reported an error regarding bandwidth value.
Btw, I ran the Apple HLS analyzer and it reported an error regarding bandwidth value.
Mhey, is stream always charged per 1000 mins? Is it not possible to charge per minute (maybe for large use cases)?
PI've seen a demo on twitter that looks really cool: https://x.com/kwindla/status/1815892810584723609 it uses daily.co on the backend to enable a webRTC connection and https://github.com/pipecat-ai/pipecat (python) to handle the pipeline of speech-to-text -> LLM -> text-to-speech and function calling etc. The webRTC stuff helps makes streaming audio really really fast obviously and that's kind of the killer feature I'm after, since it needs to be as seamless as possible. I'm trying to get my head around how to build something like this perhaps with cloudflare calls? I'd probably still use grok for the llm/whisper api calls simply because it's so fast, but I do currently use a cloudflare worker for whisper. Like in a cloudflare only setup should I be thinking about connecting the client and a durable object to a call and having the durable object orchestrate the whisper -> llm -> tts, is that possible? Any help or thoughts appreciated!
Very, very fast voice bots. Llama 3.1 running on @GroqInc.
500ms voice-to-voice response times
500ms voice-to-voice response times
7/23/24, 11:32 PM
GitHub
Open Source framework for voice and multimodal conversational AI - pipecat-ai/pipecat
HFor Downloading Videos, does it give you the option to download it in the originally uploaded format? Or do you need to separate store it somewhere else if you want to ensure you have the original?
DHi @mickiebetz any chance we can get someone to assist? It’s been nearly 5 days since we opened a ticket…. This stream state is incorrect and I’m without any way to fix TIA
MO, so if i consume lets say 5mins, i only get charged 5 mins? Thought they had to be bought in packs
Estorage is prepaid and bought as a pack, the data streamed is postpaid based on usage
MGot it, thanks
HOk, so I think I'll upload to R2 first, then pass a presigned URL for Stream to ingest
HIf that's the way
CHello, I'm having an issue with buffering for some players in the EU, is this the right place to discuss it?
DTicket 3350162. Thank you!
IHello.
I am experiencing issues with my HLS streaming setup. Specifically, we encountered problems with the 2-stream_t1_r480001.m3u8 playlist where a particular segment file (seg_835_part_all.mp4) could not be retrieved, resulting in an error. After switching to another playlist (3-stream_t1_r360001.m3u8), we faced a similar issue with a different segment file (seg_836_part_all.mp4). Finally, switching to a third playlist (4-stream_t1_r240001.m3u8) also resulted in errors.
Is this the right channel to discuss and seek help for these issues?
I am experiencing issues with my HLS streaming setup. Specifically, we encountered problems with the 2-stream_t1_r480001.m3u8 playlist where a particular segment file (seg_835_part_all.mp4) could not be retrieved, resulting in an error. After switching to another playlist (3-stream_t1_r360001.m3u8), we faced a similar issue with a different segment file (seg_836_part_all.mp4). Finally, switching to a third playlist (4-stream_t1_r240001.m3u8) also resulted in errors.
Is this the right channel to discuss and seek help for these issues?
1I've got a few stream recordings that are refusing to generate download links. Have tried turning off and back on. Anything else I should look at
Video IDs
b5b15c759d14c0be9dedfb32c2eb45d0
773482a503b2a70b15d3c2a5b859b1c6
Video IDs
b5b15c759d14c0be9dedfb32c2eb45d0
773482a503b2a70b15d3c2a5b859b1c6
EAny word on when Media Over QUIC relays will be supported by Stream? I am building out a MOQ CDN and would love to use Cloudflare. I ported the MOQ Directory API to a CF Worker with a Durable Object and I am now setting up a CF Load Balancer to handle the routing/steering for my CDN nodes.
EMHow scalable is cloudflare calls? Can I just use that instead of webrtc livestream and just block some users from using their mic/camera?
KVideo clipping feature (https://developers.cloudflare.com/stream/edit-videos/video-clipping/) is completely broken: "The video failed to be encoded due to an unknown cause."
So far have not been able to clip a single video
So far have not been able to clip a single video
Cloudflare Docs
With video clipping, also referred to as “trimming” or changing the length of the video, you can change the start and end points of a video so viewers …
II have several questions, the minimum latency is 10 seconds in a live video with respect to the person who is watching it but there are times when it reaches 25 seconds and that's a lot, can't it be lowered?
Because 1 minute of video for 5 users, it is consumed 5 minutes if they have consumed the same fragment? it is not stored in the CDN.
My unused minutes are stored for the next month?
In addition I think it is interesting to have a minimum free layer to test for example 1 or 2 hours, pay 5 dollars for 20 minutes streamed.
Because 1 minute of video for 5 users, it is consumed 5 minutes if they have consumed the same fragment? it is not stored in the CDN.
My unused minutes are stored for the next month?
In addition I think it is interesting to have a minimum free layer to test for example 1 or 2 hours, pay 5 dollars for 20 minutes streamed.
Dis the standard workflow for a solution like mine as follows:
Solution - encoding a 3D visualization on a cloud server on a per user basis (like game streaming)
Solution - encoding a 3D visualization on a cloud server on a per user basis (like game streaming)
- i detect a user wants to connect in my client (a webview desktop app or a browser)
- i tell my backend to start a stream
- my backend calls CF to get an endpoint URL, basically provision a stream for my backend
- my backend returns the CF stream URL to the client
- my 3d app instance starts up, reads the CF URL to connect my encoding session to and makes the connection
- my client app then uses the URL to create a MediaStream JS object and passes that to a video element in my client app
Dok, thanks. yep, i'm using webRTC
Ghi, some thumbnails and videos are not showing up for streamer's recorded vods. See
https://mage.stream/channel/88
https://mage.stream/channel/88
MHello, I would like to know if I can stream my live video not from OBS, but from a video I'm recording using a code I have
basically I'm checking if I can replace my current websocket which sends the video from my server to the clients in the browser with Cloudflare's Stream service, is this possible or CF stream is built for different cases?
Thanks
basically I'm checking if I can replace my current websocket which sends the video from my server to the clients in the browser with Cloudflare's Stream service, is this possible or CF stream is built for different cases?
Thanks
Nhey, does webrtc output work when input is rtmp? Im streaming video to the RTMPS endpoint and want to get the least possible latency. When using the RTMPS playback URL I get around 6 seconds
GNo the videos arent playing either. This is using rtmps.
MThanks for sharing this!
I think the default solution won't work with us because of the latency!
but in the webrtc beta you're promising a less than one-second latency, can you please confirm this?
I think the default solution won't work with us because of the latency!
but in the webrtc beta you're promising a less than one-second latency, can you please confirm this?
Aheyoo 
I am setting up direct creator uploads and it's all looking good, however I only ever seem to get the first chuck on the file uploaded, every
I am setting up direct creator uploads and it's all looking good, however I only ever seem to get the first chuck on the file uploaded, every PATCH HTTP request gets a 500 response, but the error message (10000) is simply to try again later. I'm wondering if there's anything obvious I might be missing?
DWhen using the streams product, if I stream in an RTMP(s) feed, and want to use an RTMP(s) player, is there anyway to pass what bitrate I want? For example, if I'm using FFMPEG or similar to playback the stream, being able to select the stream quality would be nice.
DI don't think you can output/read a rtmp feed, its rtmp in only.
DCan I set bitrate with HLS?
DIt would be cool if any stream traffic coming from cname would be sent via webhook/api to my app for auth - instead of forcing us to create inputs for each stream. I have 2,500 end points for customers and it will be a pain replacing each stream key.
DRTMP(s) does do playback
IRTMP output is direct passthrough, there's no encoding done by Stream when outputting to RTMP (only to HLS). So whatever you put in is whatever you get out.