Transcription failed again

[wuzimi]

Hi, could you pls help me on this error? It worked well previously. The information is here.

[ALEKS0805]

I have the same error since yesterday

[ALEKS0805]

@wuzimi did you find the fix ?

[wuzimi]

No. ALEKS0805 ***@***.***> 于 2023年5月10日周三 22:17写道：

…

@wuzimi <https://github.com/wuzimi> did you find the fix ? — Reply to this email directly, view it on GitHub <#145 (reply in thread)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/A4ZON2BJ2CH6QA56NHWRDXTXFOPQFANCNFSM6AAAAAAX4LV5FM> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[jjmaldonis]

Hey @wuzimi and @ALEKS0805, thanks for the reports. I tried to reproduce the issue but wasn't able to, so we need some more info to track down what's going on.

Can you please upload the audio you're using so I can see if it is related to the audio file itself? If you don't want to upload the audio file here, you can submit it via this form - please mention this github issue in your message so I can find your audio file.

Also, can you provide the (python?) code you're using? It looks like you're using Deepgram's Python SDK, so I tried to replicate your issue using the code below. The code below ran successfully, so there must be a more nuanced problem that we need to track down.

from deepgram import Deepgram  # pip install deepgram-sdk
import asyncio
import json
import os


# Initialize the Deepgram SDK.
DEEPGRAM_API_KEY = os.environ["DEEPGRAM_API_KEY"]  # Your Deepgram API Key
deepgram = Deepgram(DEEPGRAM_API_KEY)


async def main(url: str):
    r = await deepgram.transcription.prerecorded(
        source={"url": url},
        options={
            "model": "whisper-large",
            "punctuate": True,
            "diarize": True,
            "language": "en-us",
        },
    )

    print(json.dumps(r["results"], indent=2))


if __name__ == "__main__":
    url = "https://res.cloudinary.com/deepgram/video/upload/v1663090404/dg-audio/AI_Show_afpqeh.m4a"
    asyncio.run(main(url=url))

[wuzimi]

This is solved. Thanks! Jason Maldonis ***@***.***> 于 2023年5月18日周四 22:53写道：

…

Hey @wuzimi <https://github.com/wuzimi> and @ALEKS0805 <https://github.com/ALEKS0805>, thanks for the reports. I tried to reproduce the issue but wasn't able to, so we need some more info to track down what's going on. Can you please upload the audio you're using so I can see if it is related to the audio file itself? If you don't want to upload the audio file here, you can submit it via this form <https://help.deepgram.com/hc/en-us/requests/new> - please mention this github issue in your message so I can find your audio file. Also, can you provide the (python?) code you're using? It looks like you're using Deepgram's Python SDK, so I tried to replicate your issue using the code below. The code below ran successfully, so there must be a more nuanced problem that we need to track down. from deepgram import Deepgram # pip install deepgram-sdkimport asyncioimport jsonimport os # Initialize the Deepgram SDK.DEEPGRAM_API_KEY = os.environ["DEEPGRAM_API_KEY"] # Your Deepgram API Keydeepgram = Deepgram(DEEPGRAM_API_KEY) async def main(url: str): r = await deepgram.transcription.prerecorded( source={"url": url}, options={ "model": "whisper-large", "punctuate": True, "diarize": True, "language": "en-us", }, ) print(json.dumps(r["results"], indent=2)) if __name__ == "__main__": url = "https://res.cloudinary.com/deepgram/video/upload/v1663090404/dg-audio/AI_Show_afpqeh.m4a" asyncio.run(main(url=url)) [image: image] <https://user-images.githubusercontent.com/2326898/239266684-9384f49a-316f-4ccd-8d63-23a0d48f7314.png> — Reply to this email directly, view it on GitHub <#145 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/A4ZON2ELFIG6RVIAZ5ISWKDXGYZWLANCNFSM6AAAAAAX4LV5FM> . You are receiving this because you were mentioned.Message ID: ***@***.***>