Simplified videos on AI, LLMs, Natural Language Processing (NLP) and data analytics π
youtube.com/@analyticsCamp/videos?sub_confirmationβ¦
NLP girl (PhD) turned AI educator and developer, teaching:
-- AI agents: agentic systems and workflows π€
-- LLMs: How Large Language Models and neural networks work in Artificial Intelligence systems,
-- AI for business: AI automation and applications,
-- Generating passive income with programming and AI,
-- NLP: Using text processing machine learning methods to solve real-life problems,
-- Data science fun projects and ideas with Python and R,
And many more tips and tutorials
DM or email me for collab or sponsorship π
Analytics Camp
Making text to speech sound genuinely human instead of robotic comes down to handling rhythm and pauses correctly. Kokoro TTS offers a fantastic free option for podcast style generation that rivals paid software. I tested out a simple trick to make the output sound natural, and if you're a new sub, you can watch the tutorial here π youtu.be/nwELsTabELSM
#TextToSpeech #Python #AI #TTS #VoiceAI #Kokoro
7 hours ago | [YT] | 2
View 0 replies
Analytics Camp
Synthesising natural speech used to require massive studio setups and expensive proprietary tools. With modern voice engineering models like Maya1, you can design expressive, emotional audio right from your desktop. If you're a new sub, I put together a hands-on tutorial showing how I built a custom voice model step by step π https://youtu.be/naDDr_ABU4M
#VoiceAI #TextToSpeech #AI #Maya1 #TTS
2 days ago | [YT] | 2
View 0 replies
Analytics Camp
We are often told that chaining ten different AI agents together will automatically solve complex problems. Google research recently showed that adding more agents can actually increase hallucinations and noise if the routing logic is poor. I looked into why less is often more when designing agentic workflows. For the new subs whoβve missed the video, please watch it here π https://youtu.be/tzloGPymubI
#AIAgents #AI #Google #research #agenticAI
5 days ago | [YT] | 2
View 0 replies
Analytics Camp
Andrej Karpathy recently shared some harsh truths about current software development trends, and honestly, he makes a very fair point. Building wrappers around basic application programming interfaces is not going to cut it long term. I looked into his breakdown of what skills actually matter right now, and if you've just joined this channel, welcome and you can watch it here π https://youtu.be/81rjYo-c0-4
#SoftwareEngineering #AI #VibeCoding #AIengineer #AgentEngineering #AndrejKarpathy
1 week ago | [YT] | 4
View 0 replies
Analytics Camp
DeepSeek architecture updates always generate a lot of noise, but looking past the headlines reveals some genuinely clever engineering choices. I spent some time dissecting their model setup so you do not have to read a forty-page white paper on a Friday night. For the new subs who might have missed it, you can catch the ten-minute walkthrough of DeepSeek V4 model here π https://youtu.be/fbd6W_QIlVM
#DeepSeekV4 #paper #AI #LLMs #research
1 week ago | [YT] | 8
View 0 replies
Analytics Camp
If you're wondering what happened to me in those 2 absent months, I was trying to automate my YouTube voice with a FREE AI model locally with limited VRAM; finally, I managed to document my journey so you know what to avoid for your own channel or app π https://youtu.be/bWX7PCQVhgQ
#ai #youtubeautomation #voicecloning #texttospeech #tts #voiceai #freeai
1 week ago | [YT] | 3
View 0 replies
Analytics Camp
Not everyone has access to high-end hardware for local model training, but you can still fully fine-tune models effectively using local workflows with CPU optimisation tricks. For the new subs who have missed this, I put together an accessible guide showing how to run the code without needing an expensive GPU setup π https://youtu.be/oG0jsMVTg9w
#finetuning #LLMs #Python #AI #CodeTutorial
1 week ago | [YT] | 3
View 0 replies
Analytics Camp
Parameter-efficient fine-tuning ( PEFT) has completely changed how developers adapt open-source models for custom downstream tasks. This video is a straightforward introduction and code tutorial of LoRA without getting lost in configuration errors. For the new subs who havenβt seen it, you can check it out here π https://youtu.be/aj1V9_5nAfo
#lora #finetuning #peft #Python #MachineLearning #AI #CodeTutorial
2 weeks ago | [YT] | 3
View 0 replies
Analytics Camp
Fine-tuning large language models used to require a cluster of enterprise graphics cards costing thousands of dollars. Using QLoRA quantisation allows you to train models efficiently on a single consumer GPU without losing performance. For the new subs who might have missed this, I walked through the full implementation code in this tutorial π https://youtu.be/QzsjHUhZZLE
#FineTuning #LLM #Python #qlora #LLMfinetuning
2 weeks ago | [YT] | 3
View 0 replies
Analytics Camp
Kimi K3 recently arrived and started crushing western benchmarks, which is interesting to watch unfold in the global AI landscape. Of course, we have to look past the corporate PR and see what it actually means for developers. I broke down the details and where things are heading in this video π https://youtu.be/SugL0FRIBmk
#KimiK3 #BestAI #ArtificialIntelligence #TechNews #LLMs
2 weeks ago | [YT] | 3
View 0 replies
Load more