Failed to generate audio

[Tobe2d]

I saw the demo on the readme page and sounds very good, so I went ahead I tested the code however I run into an issue where it worked for generating one of the two with error for the 2nd one:

Failed to generate audio for W. L. Oxley: Client error '400 Bad Request'

I tested it with multiple voices and always the error for the 2nd person. even I try switching the voice ID and still the same, error for the 2nd voice.

[Tobe2d]

after restarting the installation it works but stuck on this step:

Crossfade duration: 10 ms
Crossfade duration: 10 ms

[vatsalsaglani]

I saw the demo on the readme page and sounds very good, so I went ahead I tested the code however I run into an issue where it worked for generating one of the two with error for the 2nd one:

Failed to generate audio for W. L. Oxley: Client error '400 Bad Request'

I tested it with multiple voices and always the error for the 2nd person. even I try switching the voice ID and still the same, error for the 2nd voice.

You need to add and select the voice you want in your ElevenLabs dashboard and provide those voice ids in the prompt. Also I'm not sure if the free trial will allow you to generate one short podcast.

[Tobe2d]

Thanks for the reply. I have a subscription with ElevenLabs, and after entering the Voice ID, it seems to work initially. However, it gets stuck on the final step in the log, and the files never combine.

The API successfully generates the files on ElevenLabs, but I can't get the process to complete as shown in your example.

Am I missing something, or is this a bug?

[vatsalsaglani]

@Tobe2d not exactly sure why that's happening. If you received two the Crossfade duration: 10 ms twice on the terminal then probably export is taking time. Is there any ffmpeg related exception or any other exception?