Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

google multispeaker not working #221

Open
valtahomes opened this issue Jan 1, 2025 · 14 comments
Open

google multispeaker not working #221

valtahomes opened this issue Jan 1, 2025 · 14 comments

Comments

@valtahomes
Copy link

Hi,

The other tts works pretty well like openai, elevenlabs etc. but the google multispeaker model keeps spitting out the same error (i have set up google config following the steps):

RuntimeError: Failed to generate audio: 403 Requests to this API texttospeech.googleapis.com method google.cloud.texttospeech.v1.TextToSpeech.SynthesizeSpeech are blocked. [reason: "API_KEY_SERVICE_BLOCKED"
domain: "googleapis.com"
metadata {
key: "service"
value: "texttospeech.googleapis.com"
}
metadata {
key: "methodName"
value: "google.cloud.texttospeech.v1.TextToSpeech.SynthesizeSpeech"
}
metadata {
key: "consumer"
value: "projects/22515376012"
}
metadata {
key: "apiName"
value: "texttospeech.googleapis.com"
}
, locale: "en-US"
message: "Requests to this API texttospeech.googleapis.com method google.cloud.texttospeech.v1.TextToSpeech.SynthesizeSpeech are blocked."
]
@valtahomes
Copy link
Author

i fixed the local credential issue but i got another error:

File ~/podcastfy/podcastfy/tts/providers/geminimulti.py:311, in GeminiMultiTTS.generate_audio(self, text, voice, model, voice2, ending_message)
309 except Exception as e:
310 logger.error(f"Failed to generate audio: {str(e)}", exc_info=True)
--> 311 raise RuntimeError(f"Failed to generate audio: {str(e)}") from e

RuntimeError: Failed to generate audio: 403 Multi-speaker voices are only available to allowlisted projects.

How to get the projects allowed by google gemini? Has anyone met similar issues?

Thanks.
V

@Siboooo
Copy link

Siboooo commented Jan 3, 2025

Got the same error, the service page says "This feature is only available to projects in allowlist. Please contact us if you want to use this feature." (https://cloud.google.com/text-to-speech/docs/create-dialogue-with-multispeakers). Online chat with the team will help you with that.

@nkenna
Copy link

nkenna commented Jan 3, 2025

conversation_config = { "roles_person1": "storyteller", "roles_person2": "audience participator", "podcast_name": "Echeta Podcast", "podcast_tagline": "Your Everyday and Everything Podcast", "creativity": 1, "dialogue_structure": ["scene setting", "character introduction", "rising action", "climax", "falling action", "resolution"], "conversation_style": ["narrative", "suspenseful", "descriptive", "humorous"], "engagement_techniques": ["cliffhangers", "vivid imagery", "audience prompts"], "user_instructions": "Speakers should introduce the podcast episode. Speakers should often overlap in their sentences. The conversation should be about the relationship between this story and life. Person2 should make connections with her own life!. Person1 should add a little humour to the discussion. Questions should be asked towards the end about the disscussion and every day life. The conversation should end with request to comment and advert that Echeta is available on web, apple app store and google play store." #"user_instructions": "Speakers should often overlap in their sentences. The conversation should be about the relationship between this story and life. Person2 should make connections with her own life!" }

Do you think it might be coming from the user instructions?

@santosh-kumar-g-cloudambassadors
Copy link
Contributor

Hey @valtahomes ,
As the Error Message suggests RuntimeError: Failed to generate audio: 403 Multi-speaker voices are only available to allowlisted projects., you'll need to contact Google Cloud Support Team to get Multi-Speaker voices enabled.
Navigate to Support in your GCP Project and raise a ticket under Cases.
Example -
image
They should be able to help.

@souzatharsis
Copy link
Owner

souzatharsis commented Jan 5, 2025 via email

@JamesHuckle
Copy link

Greatly appreciate the detailed reply, Santosh!

@spirillen spirillen mentioned this issue Jan 17, 2025
Closed
@flatpackfan
Copy link

Is there a way to contact Google support that doesn't require a paid support license?

@santosh-kumar-g-cloudambassadors
Copy link
Contributor

Is there a way to contact Google support that doesn't require a paid support license?

@flatpackfan No. If you're on a free tier or using Google Cloud's free credits, support is limited. But contacting support via the GCP project's "Cases" section is typically for paid support plans.

@flatpackfan
Copy link

@flatpackfan No. If you're on a free tier or using Google Cloud's free credits, support is limited. But contacting support via the GCP project's "Cases" section is typically for paid support plans.

Yeah, that was my conclusion as well. So when you're on the free tier there's basically no way to enable multi speaker tts support right?

(would be good to mention this in the documentation)

@souzatharsis
Copy link
Owner

souzatharsis commented Jan 31, 2025 via email

@santosh-kumar-g-cloudambassadors
Copy link
Contributor

santosh-kumar-g-cloudambassadors commented Jan 31, 2025 •
edited
Loading

@flatpackfan No. If you're on a free tier or using Google Cloud's free credits, support is limited. But contacting support via the GCP project's "Cases" section is typically for paid support plans.

Yeah, that was my conclusion as well. So when you're on the free tier there's basically no way to enable multi speaker tts support right?

(would be good to mention this in the documentation)

Yes @flatpackfan that's correct. Google TTS multi-speaker support is a premium feature, so it requires a paid account to enable. Unfortunately, it's not available on the free tier.

@flatpackfan
Copy link

Yes @flatpackfan that's correct. Google TTS multi-speaker support is a premium feature, so it requires a paid or billed account to enable. Unfortunately, it's not available on the free tier.

Correct me if I'm wrong but having a billed account isn't enough (I have that). You also need paid support.

@santosh-kumar-g-cloudambassadors
Copy link
Contributor

Would love to have some of you to open a PR with updated documentation
given your experience on this issue.

Perhaps update this one:
https://github.com/souzatharsis/podcastfy/blob/main/usage/config.md

Best Regards,
…

@souzatharsis Done ✅ - #235

@guiblachon
Copy link

Hello,

FYI, the allowlist is closed on google side, what they told me :

The request was denied because they were no longer accepting them, and since this experimental product is not ready for public use. They are currently developing a new version of multi-TTS and are therefore not accepting new access requests as of January 30th.

Solution/Recommendation Provided: Is to check back to see when this new product is available for multi TTS.

Conclusions: The allow list has been closed and we will have to wait until the new product is released.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
8 participants
@souzatharsis @JamesHuckle @nkenna @Siboooo @guiblachon @valtahomes @santosh-kumar-g-cloudambassadors @flatpackfan