I downloaded native data and i noticed - compared to other data the manifest.json and manifest.json.crc are missing. Does this influence anything of my data?

(for a Neon Recording)

Hi everyone. I am currently using Neon Player and have a few questions regarding the usage. Surface definition: I want to define a screen as a surface, while the player does recognize the apriltags, it is very laggy when I try to move the position of the markers or does not work at all. Do I have to change something in regards to height and width? I already have downloaded the newest version so it cannot be that. Events: i have commuicated certain time stamps to the eye tracker. However I can only see this time stamp when being in the exact time frame. Is there another way to jump between events other than the annotation time line?

Thank you for your quick reply. Yes, I do not have an overview of the events (or better, I can't find one) 🙂

There is a little tab that allows this.

Is it possible to programmatically pull in the eye videos using the v2 api?

Hi @user-3ff3c1 , the Pupil Cloud API does not currently provide that level of granularity. It is necessary to pull the whole recording file over the API and then extract the eye video from the resulting ZIP file.

If you'd like to see this added, please make a 💡 features-requests .

Is there a way to use an API to automatically export recordings from pupil_player? We have the 000 folder and can open pupil_player to export, but we have hundreds of files and want to automate that process instead of opening the gui. I can't find any documentation on the website. Thanks for any help

Hi @user-3ff3c1 , it sounds like you are using the Timeseries CSV endpoint. You rather want to use the Native Recording Data endpoint.

Hi @user-3ff3c1 , apologies for the delay.

If you just want the eye videos, then you can use this endpoint:

https://api.cloud.pupil-labs.com/v2/workspaces/{WORKSPACE_ID}/recordings/{recording_id}/files/{filename}

where filename can be replaced with eye.mp4.

You could wrap it up in a Python function like this:

import requests

def get_individual_file(filename):
    url = f"https://api.cloud.pupil-labs.com/v2/workspaces/{WORKSPACE_ID}/recordings/{recording_id}/files/{filename}"
    response = requests.get(url, stream=True, headers={"api-key": API_TOKEN})

    save_path = f"{filename}"
    chunk_size = 128

    with open(save_path, "wb") as fd:
        for chunk in response.iter_content(chunk_size=chunk_size):
            fd.write(chunk)

    return response.status_code

And then call it as follows:

def get_eye_video(recording_id):
    get_individual_file("eye.mp4")

Hello, I am using the latest Pupil Labs Realtime API for Python. I am enjoying the recently added audio support. Upon integrating audio support into my code base, I couldn’t find a reliable method of detecting when the microphone is muted in the Companion mobile application. Thus, I implemented a timeout on the audio queue to detect this case. • Am I correct that there isn’t a way to reliably detect the microphone is muted in the Companion mobile app? • Looking at models.py, it appears as though if the world sensor is connected, the audio sensor connected status is unconditionally set to True. • If my assumptions are true, can a check be added to the API that reports the microphone status or the connected status of the audio sensor be updated accordingly?

To add a bit to my colleague's response, the audio is multiplexed on the video RTSP, but on our client we generate a separate stream as streaming only audio is a valid scenario.

The microphone status is not unconditionally set to True, but as audio has not it's own RTSP stream and the status does not report the mic's , we check the SDP of the RTSP to check whether it contains the audio signal.