audio-analysis 4

Gemini can pass the bar exam and solve maths olympiad puzzles. But it’s music-deaf. nitin kumar asked Gemini to rate 40 songs on joy, sadness, tension, nostalgia, etc. and compared it with human ratings. There was ZERO correlation between the two. It’s like it’s a different species. In fact, if you just predict the average emotion for every single song, you’d still do 1.2× to 2× better than Gemini! It wasn’t adding noise to a signal. It was subtracting subtracting signal from noise! In fact, for one song, the correlation was -88%, i.e. it predicted the exact opposite emotions. ...

Transcribe call recordings guessing speaker names using the latest Gemini Pro model on Google AI Studio. Append all speakers, and who spoke when, for context. Transcribe this call recording with Anand (LLM expert, Straive/Gramener). DO NOT MISS ANY PART OF THE CONVERSATION. Drop verbal tics and fillers (um, uh, etc). Correct spelling and grammar but otherwise don't modify the original words. Add English translations to any non-English parts. Mark inaudible or unclear segments as "[inaudible]". Mark uncertain words with like "[word?]" or ambiguous possibilities like "[word1? word2?]". Break it into LOGICAL paragraphs, each paragraph with a **Speaker**: [Timestamp] content ...., e.g. **Anand**: [00:13] When did ... Guess speaker names. If unsure, use **Unsure**: ... **Make key points / takeaways / memorable statements bold**. **I repeat: Transcribe EVERY part of the conversation. Don't miss any turns.**

Wikipedia says the human voice frequency for speech is between 85 to 155 Hz for men, and 165 to 255 Hz for women. That set me thinking. What is the limit to our hearing? How do sounds differ? How can we synthesise speech? What are the limits to our hearing? Kids can hear frequencies from 20 Hz to 20 kHz, while adults hear only up to 12-14 kHz (Frequency Range of Human Hearing). ...

It’s possible to make sense out of keyboard whispers.