api.video lets you quickly and securely deliver on-demand and live-stream videos directly from your website, software, or app. 

api.video

Glossary-Rate-limit.svg

## What is an API rate limit?
An API rate limit is a threshold placed on the number of API calls that can be placed over a period of time.  For example: you may only mayke 3 login attempts an hour, or make 150 calls to Twitter every 5 minutes.

API services have upper limits on the number of requests that can be handled every second (millisecond).  If one user (or several users) begin taking a substantial portion of the available bandwidth of the API - the API will fail to work for others,  By placing limits on individual users, API developers can protect their infrastructure from too many requests.

Rate limits can also affect bandwidth. For those in the USA, your "unlimited 4G data" mobile data plan is typically 25 or 40 GB at 4G speeds, and then you are rate limited - your plan is throttled to 2g for the rest of the month.

thumbnail_ratelimit.png

ratelimit.png

## Rate limits at api.video

We currently have no rate limits on our API, but we do monitor for accounts that are using excessive numbers of calls, and will work with that developer to mitigate their usage.  We are considering rate limits on video uploads, but these are still under development.

### Rate limits on video playback

There are no limits on video playback. If there was ever an unprecedented global "viral video" at api.video, we would be able to handle the load.  On the first request from our backend, api.video would deliver the video to our global video CDN, where it would be cached.  The delivery from our CDN will be fast an efficient, and since it is abstracted from our infrastructure - will not impact and other video playback from api.video.

Glossary-RTMP.svg

## What is RTMP?
RTMP stands for Real Time Messaging Protocol and for many years, it's been the standard for streaming video over the internet. What really put RTMP on the map was Adobe's Flash Player. Flash became so popular that browsers and encoders everywhere rushed to support it, which included support for RTMP. Because of this, today most encoders and media servers can support RTMP. That's why it's still the standard for streaming video and live streaming. 

The RTMP streaming protocol is TCP-based (Transmission Control Protocol) and designed to maintain constant, low-latency connections between a video player and server. The design allows RTMP to provide smooth streaming for viewers. And because RTMP uses TCP, it's very reliable. TCP uses a three way handshake when transporting data - the client contacts the server, the server accepts and responds, then the client acknowledges the response. At this point a session is opened, and RTMP can start streaming video. 

RTMP was popular up through the early 2010s, but then came the rise of HLS (created in 2009) and adaptive bitrate streaming. Adaptive bitrate streaming allows you to optimize content based on available connectivity and viewer devices. This type of streaming requires the use of HTTP-based protocols. So today, RTMP is used to move between servers, but when it's time to deliver content to the viewer, the last mile of delivery usually uses HLS and adaptive bitrate delivery. 

RTMP will likely continue to be popular because so many encoders and servers still support it, but over time, other protocols will replace it. If you want more details about the benefits and drawbacks of RTMP, check out api.video's blog post: [What is RTMP and Why Do We Use it for Live Streaming?](https://api.video/blog/video-trends/what-is-rtmp-and-why-do-we-use-it-for-live-streaming/)

For Live streaming, check this article [comparing RTMP to SRT](https://api.video/blog/video-trends/srt-vs-rtmp/).

Glossary-RTMPS.svg

## What is RTMPS (Real-Time Messaging Protocol Secure) ?
RTMPS, which stands for Real-Time Messaging Protocol Secure, is an extended version of the [RTMP (Real-Time Messaging Protocol)](https://api.video/what-is/rtmp-real-time-messaging-protocol/) that adds a layer of security through SSL/TLS encryption. It is primarily used for streaming live audio, video, and data over the Internet. RTMPS ensures that the content being streamed is protected from eavesdropping and tampering during transmission.

api.video supports RTMPS, if you are interested, check the blog article about [how to set up RTMPS using api.video](https://api.video/blog/product-updates/rtmps-livestreaming/), you can either start your livestream with RTMPS through the dashboard or through the API.

## Key Components of RTMPS
At its core, RTMPS builds upon the RTMP base, adding a crucial SSL/TLS layer for encryption and security. The protocol employs a handshake process to establish a secure connection between the client and server. Once established, it uses chunking to break data into manageable pieces for transmission and multiplexing to allow multiple streams of data to be sent simultaneously.

The RTMPS process begins with the client initiating a connection to the server over SSL/TLS. This is followed by an SSL/TLS handshake, where the client and server establish a secure encrypted connection. Within this secure tunnel, the standard RTMP handshake then occurs. After successful handshakes, one or more streams are created for data transmission. Audio, video, and other data are then sent in real-time through this encrypted connection. Either party can securely close the connection when streaming is complete.

## Applications and Use Cases
RTMPS finds wide application in various streaming scenarios. It's commonly used for live video broadcasting, enabling secure streaming of events, webinars, or personal broadcasts. In video conferencing, it ensures private and secure communication for real-time video calls. The gaming industry utilizes RTMPS to protect live game streams from potential security threats. In the education sector, it secures live educational content delivery to students in e-learning environments. RTMPS also plays a role in IoT device communication, enabling secure real-time data transmission from Internet of Things devices. Additionally, it facilitates secure internal broadcasts within organizations for corporate communications.

## Advantages and Considerations
The primary advantage of RTMPS is its enhanced security, protecting sensitive content from unauthorized access during transmission. It's also highly compatible with existing RTMP infrastructure, requiring minimal changes for implementation. RTMPS maintains the low-latency benefits of RTMP while adding a layer of security.

However, there are considerations when implementing RTMPS. The encryption process can add some computational overhead, which may impact performance in resource-constrained environments. Proper management of SSL/TLS certificates is crucial for maintaining security. Additionally, specific firewall configurations may be necessary to allow RTMPS traffic, which can be a consideration in some network environments.

## Future Prospects
As video streaming continues to grow in importance, RTMPS is likely to evolve. We can expect to see integration with new codecs, allowing it to work efficiently with emerging video and audio compression technologies. Efforts to enhance performance and reduce the overhead of encryption for even lower latency are likely to continue. As security becomes increasingly critical, broader adoption across various platforms is anticipated.

However, it's worth noting that while RTMPS is currently widely used, newer protocols like [SRT (Secure Reliable Transport)](https://api.video/blog/product-updates/SRT-livestreaming/) may eventually supersede it in some applications. Nonetheless, RTMPS remains a crucial protocol in the world of real-time video streaming, balancing the needs for security, low latency, and compatibility with existing systems.

Glossary-run-length-encoding.svg

## What is run-length encoding?
Run-length encoding (RLE) is a lossless compression method where sequences that display redundant data are stored as a single data value. This value represents the repeated block, and shows how many times it appears in the image. During decompression, the image can be reconstructed exactly from this information.

This type of compression works best with simple images and animations that have a lot of redundant pixels. It is useful for black and white images in particular. For complex images and animations that do not have many redundant sections, RLE can make the file size bigger rather than smaller. Therefore it is important to understand the content and whether this algorithm will help or hinder compression.

A graphic showing how run-length encoding works by compressing a string of the same letter down to 1 of the letter and a number representing the number of times the letter appears in a row.

run-length.png

## History of run-length encoding
This technique was first patented by Hitachi in 1983, and was regularly used when transmitting analog television signals, even as early as 1967. It is less popular today because there are other more advanced options available. You can still find it in use for color fax machines, icons, line drawings, and simple animations. The method is also used in TIFF and PDF files.

## Run-length encoding and LZ77
LZ77 is an algorithm that compresses data in a similar way to RLE. It replaces repeated occurences of data with a reference to a single copy of that data that exists earlier in the uncompressed stream. If there is a match to a chunk of data, then LZ77 encodes a match that represents the distance between the two chunks of data and the length of each chunk. To find matches, LZ77 has to track a chunk of recent data, which can be of varying sizes. The bigger the chunk of recent data, the bigger a chunk it can search through for a match. LZ77 is also called sliding-window compression because it is viewing a chunk of data that can be of varying sizes to find matches. Pairs can repeat multiple times, which makes LZ77 similar to RLE, just a bit more complex.

The concept used to create RLE also appears in LZ78, and other popular compression methods we still use today, like GIF and DEFLATE. The DEFLATE algorithm appears in PNG and ZIP.

## Drawbacks of run-length encoding
The main drawback of RLE is that it is best suited for simple and repetitive data only. Here are a few more aspects to keep in mind when working with RLE:
- The original data is ot instantly accessible, you must decode everything before you can access anything.
- You cannot tell how large the decoded data will be. This can cause problems when you have limited space to decompress the file in. In this case, you can either store the size of the original data as metadata, or use this compression method with a dynamic buffer and enough space to decompress the content in.

&nbsp;  

Learn more about the history of video compression and the different algorithms that have appeared over time in our blog article about [The History of Video Compression Standards, From 1929 Until Now](https://api.video/blog/video-trends/the-history-of-video-compression-starts-in-1929/).

## What is video startup time?
Video startup time is the time from pressing "play" to the video starting to move.

When a video takes a long time to startup, viewers begin to abandon, and go do something else - maybe even abandoning the site. So it is imperative minimize the amount of time between pressing play and having the video begin playback.

## The video tag
In the video tag, you can specify [autoplay](https://api.video/what-is/autoplay/) to esure fast startup. But for videos where a user will press a button to start the video, using the [preload](https://api.video/what-is/preload/) attribute can ensure video locally available for playback.

## Adaptive streaming
api.video uses [adaptive bitrate streaming](https://api.video/what-is/adaptive-bitrate-streaming/) (ABS). ABS has multiple versions of the videos at different sizes.  One way that ABS streams can ensure fast startup is to start the video playback with a lower quality video (240p video is much smaller in KB than a 1080p video).  The first few seconds arrive very quickly, the player is able to calculate the network speed, and then subsequent segments are shown at a higher quality. Many video streaming services use this feature, as the first few seconds of a move or TV show are often a black screen anyway.

## CDN
A robust [CDN](https://api.video/what-is/content-delivery-network/) network ensures that all video segments are stored on the edge of the internet near the viewers - reducing the round trip time for each request.  The less time spent in transit is less time in delay for the video to begin on the users' device.

Glossary-Preload.svg

## What is the preload attribute in HTML5?
Preload is an attribute of the HTML5 video tag.  The preload attribute lets the developer advise the browser how much of the video should be preloaded prior to viewing.  There are 3 values for the preload attribute:
- None: preload=none tells the browser that there is a video, but to download 0 bytes of the file.  This clearly saves data, but at the cost of a delayed video startup - there is no video stored locally on the device.
- metadata: preload=metadata tells the browser to download a percentage of the video (typically 2-5% of the file) to get metadata (size, bitrate, etc. from the video, as well as have a bit of teh video ready for immediate playback.  **IF preload is not set, the default setting is metadata**
- auto: preload=auto tells the browser to download the entire video.  This is the most data intensive, but results in full playback of the video when the video playback has started.  *If the preload attribute is not assigned a value, the auto selection is used*  

&nbsp;  

> Note that ```preload=auto``` is a suggestion that the entire file can be downloaded, but the browser may have a different algorithm and download a larger percentage than ```preload=metadata``` instead of the entire file.  

&nbsp;  

The autoplay attribute takes precedence over preload. For example, in the following video tag:
```
<video src=myvideo.mp4 preload=none autoplay>
```
The browser will download the entire video.

## Preload at api.video
Preload is not typically used with streaming video, as the video player can adapt playback based on  network conditions - giving a more nuanced approach to video download.

## What is the video poster HTML5 attribute? 
The video poster is the HTML5 attribute for the video tag that defines the image that appears in front of a video before it starts playing.

If no poster tag is defined in the video tag, the browser will display the first frame of the video. Since many videos begin with a black screen, this is not visually appealing, so a different screenshot from the video, or perhaps an unrelated image is used.

The poster tag points to an image: 

```
poster=https://example.com/poster.jpg
```

There is no way to serve alternative formats with the poster tag, so typically a format that is supported in many browsers (like jpg or png) are used.  

&nbsp;  

> Note: The poster image is downloaded before the video, so if your video is going to autoplay, this will block the video from loading.  Additionally, creating a poster of the first frame of the video is redundant, as the player will do this anyway, and this results in that frame being downloaded twice.

## Poster at api.video
- The poster tag is called *thumbnail* at api.video.  
- The default thumbnail is selected from the "middle frame" of the video (example: the 15 minute point of a 30 minute video).

## Changing the thumbnail
- Set a frame from the video: The [pick a thumbnail](https://docs.api.video/reference#patch_videos-videoid-thumbnail) endpoint sends a timestamp where the image should be extracted from the video and set as the poster.
- [Upload a Thumbnail](https://docs.api.video/reference#post_videos-videoid-thumbnail): Upload a jpg image to be used as the video thumbnail.


Glossary - Poster.svg

A list of different live streams, two are news, one is Rick Astley, one is Doug

large_screenshot-2021-06-14-at-13-30-50.png

medium_screenshot-2021-06-14-at-13-30-50.png

small_screenshot-2021-06-14-at-13-30-50.png

thumbnail_screenshot-2021-06-14-at-13-30-50.png

poster-example.png

Glossary-OTT.svg

## What is Over the top (OTT)?

Over the Top (OTT) describes how a video can be delivered to customers directly over the internet.  The term derives from the delivery "on top" of basic internet protocols, bypassing the traditional providers of video: TV, cable and satellite. OTT video streaming has revolutionised the way movies and shows are watched by consumers.

There are many OTT providers, including Netflix, Hulu, Disney+ and more.  Many "traditional broadcasters have created their own OTT services (HBO Max, Peacock (NBC), etc). 

OTT delivery has led to new viewing strategies by customers. Now that shows are available on demand, consumers are no longer bound to a specific time or day of the week for a TV show. As a result, viewing patterns have changed in a way that is more convenient to the viewer. Further, the concept of "binging" a show - watching an entire season in the course of a few hours or days has also arisen from OTT video streaming.

## OTT at api.video

api.video hosts and streams video on demand.  If you have OTT content that you want delivered quickly and efficiently worldwide, we can host and deliver that content for you.  We also offer "private videos'' that only allow authenticated users access to the movies, and prevent the sharing of the video urls.


## What are Neural Networks?

Neural Networks are a fundamental component of [deep learning](https://api.video/what-is/deep-learning/), inspired by the structure and function of the human brain. In the context of video technology, neural networks process vast amounts of visual data to perform tasks such as object recognition, scene understanding, and even video generation. These networks consist of interconnected nodes (neurons) organized in layers, which work together to analyze and interpret complex patterns in video content.

## Types of Neural Networks in Video Processing
Several types of neural networks are particularly relevant to video technology:
1. Convolutional Neural Networks (CNNs): Excel at processing grid-like data, making them ideal for analyzing individual video frames.
2. Recurrent Neural Networks (RNNs): Specialize in sequential data, allowing them to process temporal aspects of video.
3. Long Short-Term Memory Networks (LSTMs): A type of RNN that can capture long-term dependencies in video sequences.
4. Generative Adversarial Networks (GANs): Used for generating new video content or enhancing existing footage.

## How Neural Networks Process Video Data
Neural networks process video data through a series of transformations:
- Input Layer: Receives raw video data, typically as a sequence of frames.
- Hidden Layers: Apply various operations (e.g., convolutions, pooling) to extract features from the input.
- Output Layer: Produces the final result, which could be a classification, a generated frame, or other video-related information.  

&nbsp;  

During training, the network adjusts its internal parameters through backpropagation, minimizing the difference between its predictions and the actual desired outputs.

## Applications in Video Technology
Neural networks have revolutionized numerous aspects of video technology. They enable video classification, automatically categorizing videos based on content, genre, or style. Object detection and tracking algorithms can identify and follow specific objects or people across video frames. Neural networks also play a crucial role in video enhancement, allowing for upscaling resolution, reducing noise, or even colorizing black and white footage.

Additionally, they are employed in video generation, creating new video content based on learned patterns or textual descriptions. Neural networks have even been applied to video compression, developing more efficient algorithms for video storage and streaming. Finally, facial recognition techniques leverage neural networks to identify and analyze faces in video footage for various applications.

## Challenges and Considerations
While powerful, neural networks in video technology face several challenges:
- Computational Requirements: Training and running complex neural networks on video data can be computationally intensive.
- Data Hunger: Neural networks often require large amounts of diverse, high-quality video data for effective training.
- Interpretability: The complex nature of neural networks can make it difficult to understand their decision-making processes.
- Overfitting: Networks may become too specialized to their training data, performing poorly on new, unseen video content.

## The Future of Neural Networks in Video
As neural network technologies continue to advance, we can expect several exciting developments. Improvements in hardware and network architectures will enable more sophisticated neural network operations to be performed in real-time, even on mobile devices. Advancements in 3D and 4D convolutions will allow networks to better understand spatial and temporal relationships in video.

Future networks will likely integrate the processing of visual, audio, and textual data more seamlessly, enabling more comprehensive video understanding through multimodal approaches. 

Additionally, self-supervised learning techniques could reduce the need for large amounts of labeled data, making it easier to train networks for specific video tasks.

Neural networks will undoubtedly offer new possibilities for content creation, analysis, and user experience. However, addressing their challenges related to efficiency, interpretability, and ethical use will be key to realizing their full potential in the video industry.

Video on demand

Live streaming

Analytics

Video infrastructure

Video player

AI features

See all features

Online learning & Corporate training

Social & Entertainment

Marketplace & E-commerce

Communication & UGC

Generative AI

Quickstart guides

Documentation

Ecosystem catalog

Clients & SDKs

Community

Blog

Demos

Customer success stories

Help center

Rate limit

What is an API rate limit?

Rate limits at api.video

Rate limits on video playback