The Spanish example (Miguel) is really bad.
it's more meant to show code switching. more examples here: https://suno-ai.notion.site/Bark-Examples-5edae8b02a604b54a4...
Bark – Text-prompted generative audio model
11–20 of 99 posts
Re: Bark – Text-prompted generative audio model
#12Any idea what the training data for this is? Looking at the model, it looks like it is literally just copy-paste from Karpathy's nanoGPT, so the training data is what's most interesting. Pretty amazing anyway.
Re: Bark – Text-prompted generative audio model
#13Well i can see it becoming sexy soon
Soon? Is the model filtered/censored?
From the readme:
>Bark has the capability to fully clone voices - including tone, pitch, emotion and prosody. The model also attempts to preserve music, ambient noise, etc. from input audio. However, to mitigate misuse of this technology, we limit the audio history prompts to a limited set of Suno-provided, fully synthetic options to choose from for each language.
It's not immediately clear how the audio history prompts are created.
Re: Bark – Text-prompted generative audio model
#14Am I hallucinating or didn't several of the examples have background audio artifacts, like it's been trained on speech with noisy backgrounds, I'm guessing audio from movies paired with subtitles? Having random background audio can make it quite hard to use in production.
Re: Bark – Text-prompted generative audio model
#15I tried it. It seems to hallucinate easily. The generated audio isn't what I provided.
It seems to be easily reproducible if I specify a non-existing speaker?
audio_array = generate_audio(text_prompt, 'en_speaker_3')
Re: Bark – Text-prompted generative audio model
#16Any idea what the training data for this is? Looking at the model, it looks like it is literally just copy-paste from Karpathy's nanoGPT, so the training data is what's most interesting. Pretty amazing anyway.
I found a secret demo page that shows in real time how they assess any sound file's mood swings along with number of detected laughs, coughs, etc. Guessing that ability is involved somehow.
Re: Bark – Text-prompted generative audio model
#17Very cool. Side note: bark-gpt.com is already taken for a dog translator: "The world’s first AI powered, real-time communications tool between humans and their furry best friends."[0] I only know this because my law firm partner's name is Bark, and I wanted to automate some legal work and name the software "Bark GPT" after him. [0] https://www.bark-gpt.com/
I want to see the training set for this
Woof!
Re: Bark – Text-prompted generative audio model
#18Earlier quoted context omitted.
Soon? Is the model filtered/censored?
From the readme: >Bark has the capability to fully clone voices - including tone, pitch, emotion and prosody. The model also attempts to preserve music, ambient noise, etc. from input audio. However, to mitigate misuse of this technology, we limit the audio history prompts to a limited set of Suno-provided, fully synthetic options to choose from for each language. It's not immediately clear how the audio history prom…
I don't know how they're made exactly, but one can just edit the code a bit and delete the restriction to just the given audio history prompts. It's literally just enforced, affect, with a simple "assert" command.
Re: Bark – Text-prompted generative audio model
#19Awesome stuff! Can’t wait to see where this company goes
Re: Bark – Text-prompted generative audio model
#20Earlier quoted context omitted.
Soon? Is the model filtered/censored?
From the readme: >Bark has the capability to fully clone voices - including tone, pitch, emotion and prosody. The model also attempts to preserve music, ambient noise, etc. from input audio. However, to mitigate misuse of this technology, we limit the audio history prompts to a limited set of Suno-provided, fully synthetic options to choose from for each language. It's not immediately clear how the audio history prom…
history prompts are just unconditionally generated TTS from the same model. any of those can be used as history, but for convenience 10 are provided for each language (to generate things with consistent voices)