search

PsychoinformaticsLab / pliers

Automated feature extraction in Python
https://pliers.readthedocs.io/en/latest/
BSD 3-Clause "New" or "Revised" License
292 stars 68 forks source link

Document limits on Google Cloud requests ? #481

Open emdupre opened 1 year ago

emdupre commented 1 year ago

In trying to include the GoogleSpeechtoTextAPI in my pliers graph, I'm receiving the following error (exact values differ by stimuli):

    raise HttpError(resp, content, uri=self.uri)
googleapiclient.errors.HttpError: <HttpError 400 when requesting https://speech.googleapis.com/v1/speech:recognize?alt=json returned "Request payload size exceeds the limit: 10485760 bytes.". Details: "Request payload size exceeds the limit: 10485760 bytes.">

This is about 10.4MB, which is less than the size of my stimulus mp4 (56.1MB), suggesting that pliers is correctly supplying only the audio track to the API. Nonetheless, I was surprised that the Google Cloud APIs would have such a small payload limit.

After looking into it, it seems that Google might have separate limits for the free trial tier. It's also possible that my requests are being handled as synchronous or streaming requests, which have a lower cap.

This threw me for a bit of a loop, so I'm not sure if it's my own user error or something worth documenting for others who might experience the same thing !

adelavega commented 1 year ago

I would recommend trying ReviAI, it's a really good transcription service. Thanks for the report