Michael Markert's picture

1

Michael Markert

mmarkert

·

AI & ML interests

None yet

Recent Activity

replied to m-ric's post 1 day ago

I've made an open version of Google's NotebookLM, and it shows the superiority of the open source tech task! 💪 The app's workflow is simple. Given a source PDF or URL, it extracts the content from it, then tasks Meta's Llama 3.3-70B with writing the podcast script, with a good prompt crafted by @gabrielchua ("two hosts, with lively discussion, fun notes, insightful question etc.") Then it hands off the text-to-speech conversion to Kokoro-82M, and there you go, you have two hosts discussion any article. The generation is nearly instant, because: > Llama 3.3 70B is running at 1,000 tokens/seconds with Cerebras inference > The audio is generated in streaming mode by the tiny (yet powerful) Kokoro, generating voices faster than real-time. And the audio generation runs for free on Zero GPUs, hosted by HF on H200s. Overall, open source solutions rival the quality of closed-source solutions at close to no cost! Try it here 👉👉 https://huggingface.co/spaces/m-ric/open-notebooklm

liked a Space 4 months ago

tonyassi/voice-clone

View all activity

Organizations

None yet

mmarkert's activity

replied to m-ric's post 1 day ago

I tried it and the results are really promising, thank you! I use NotebookLM to provide texts to university students as easily consumable podcasts so an open alternative for me is gold. I wish Kokoro or any other OS TTS model would speak German but unfortunately I did not found anything useful while NotebookLM can be forced to speak German via prompt although it is not supported officially right now. If I try this in your app the text is German but the accent is so strong that I am not able to understand anything as a native German speaker 😅.

liked a Space 4 months ago

Voice Clone

Create a personalized voice recording from text