AI Writing Resources & TTS to Video & prompts.

I am not a fabulist, so we are not competitors. To your pens, varlets!

First the free eBooks then the AI. A Festivus for the Rest of Us!

Workflow: local generation using ltxv and comfyui.

This girl does not exist. Learn Chinese

Free eBooks

MANDARIN CHINESE WORD GAMES HANZITUTION

Mastering Populism & Authoritarian Democracy

The Multistate Bar Examination (MBE) Multiple Choice Questions

The world according to Columbus

AI Writing and Fiction

If you are interested in using AI as an assistant for fiction writing these links will prove useful. It’s not at all my genre. You would be likeliest to sell books in romance, then horror, and by picking a very specific niche.

1. Plot Generator:

https://www.plot-generator.org.uk/

1. Plottr:

https://www.plottr.com/

1. Charahub:

https://charahub.com/

1. Artbreeder:

https://www.artbreeder.com/

1. World Anvil:

https://www.worldanvil.com/

1. Campfire:

https://campfiretechnology.com/

1. Reedsy Prompts: https://blog.reedsy.com/creative-writing-prompts/

2. Sudowrite:

https://www.sudowrite.com/

Text to Speech, and then Speech to video: I use piper on local install.

Either all this looks really obvious and easy or it doesn’t and if it doesn’t “there is no royal road to learning” you’ll just have to plow through it, start by getting linux ditch windows (windows is cockroaches). Then learn bash, then CAT AWK SED and then learn python, and then the fun really starts.

Workflow: I generated the transcript using notebooklm from various sources on global strategy (NOT all the ones i would have included this is an experiment). Then converted the .wav to .mp4 using the following code:

#!/bin/bash

for file in *.wav; do

    ffmpeg -i "$file" -filter_complex "[0:a]showwaves=s=1280x720:mode=line:rate=25,colorkey=black:0.01:0.1,format=yuva420p[v]" -map "[v]" -map 0:a -c:v libx264 -c:a aac -b:a 192k -shortest "${file%.wav}.mp4"

done

1. Showspectrum: This filter creates a scrolling spectrogram of the audio[5]. You can use it like this:

```bash

ffmpeg -i input.wav -filter_complex "[0:a]showspectrum=s=854x480:mode=combined:slide=scroll:saturation=0.2:scale=log,format=yuv420p[v]" -map "[v]" -map 0:a -b:v 700k -b:a 360k output.mp4

```

2. Avectorscope: This filter creates a circular visualization of the audio waveform[5]. Here's an example:

```bash

ffmpeg -i input.wav -filter_complex "[0:a]avectorscope=s=480x480:zoom=1.5:rc=0:gc=200:bc=0:rf=0:gf=40:bf=0,format=yuv420p[v]" -map "[v]" -map 0:a -b:v 700k -b:a 360k output.mp4

```

3. Ahistogram: This filter creates a scrolling histogram of the audio[5]:

```bash

ffmpeg -i input.wav -filter_complex "[0:a]ahistogram=s=hd480:slide=scroll:scale=log,format=yuv420p[v]" -map "[v]" -map 0:a -b:a 360k output.mp4

Prompt Fu.

You can use a.i. to help you learn to program. Here’s a prompt I used because I am lazy.

I am using ollama webui on local install. it allows me to import prompts. but my prompts are plaintext, and i don't know the json required. (1) can you give me a skeleton json? (2) can you write a BASH script to tranform foo.txt into bar.json where foo is a plaintext prompt? (3) can you provide a batch version of that script so i can process the entire directory ~/prompts to transform all the files therein into well-formed json prompts?

##This is the general format for a prompt in JSON
{

  "prompt": "AIM HERE BOYS. SOUPS ON COME AND GET IT!",

  "max_tokens": 0,

  "temperature": 0.0,

  "top_p": 0.0,

  "top_k": 0,

  "typical_p": 0.0,

  "repetition_penalty": 0.0,

  "num_beams": 0,

  "no_repeat_ngram_size": 0,

  "early_stopping": false,

  "length_penalty": 0.0,

  "min_length": 0,

  "max_length": 0,

  "stop_sequences": [],

  "use_cache": true

}

##THIS BASH SCRIPT convert_prompt.sh CONVERTS THE TEXT FILE foo.txt INTO THE PROMPT JSON bar.json

#!/bin/bash

# Set input and output files
INPUT_FILE="foo.txt"
OUTPUT_FILE="bar.json"

# Set default parameters (adjust as needed)
MAX_TOKENS=512
TEMPERATURE=0.7
TOP_P=0.9
TOP_K=40
TYPICAL_P=0.9
REPETITION_PENALTY=1.0
NUM_BEAMS=4
NO_REPEAT_NGRAM_SIZE=3
EARLY_STOPPING=false
LENGTH_PENALTY=0.0
MIN_LENGTH=0
MAX_LENGTH=512
STOP_SEQUENCES=()
USE_CACHE=true

# Read prompt from input file
PROMPT=$(cat "$INPUT_FILE")

# Create JSON output
cat <<EOF > "$OUTPUT_FILE"
{
  "prompt": "$PROMPT",
  "max_tokens": $MAX_TOKENS,
  "temperature": $TEMPERATURE,
  "top_p": $TOP_P,
  "top_k": $TOP_K,
  "typical_p": $TYPICAL_P,
  "repetition_penalty": $REPETITION_PENALTY,
  "num_beams": $NUM_BEAMS,
  "no_repeat_ngram_size": $NO_REPEAT_NGRAM_SIZE,
  "early_stopping": $EARLY_STOPPING,
  "length_penalty": $LENGTH_PENALTY,
  "min_length": $MIN_LENGTH,
  "max_length": $MAX_LENGTH,
  "stop_sequences": [],
  "use_cache": $USE_CACHE
}
EOF

##THIS IS THE BATCH SCRIPT TO PROCESS THE FOLDER WHERE YOU KEEP YOUR PROMPTS IN TEXT FORMAT

#!/bin/bash

# Set input directory
INPUT_DIR="~/prompts"

# Set output directory (same as input directory)
OUTPUT_DIR="$INPUT_DIR"

# Loop through all files in input directory
for file in "$INPUT_DIR"/*.txt; do
  # Get filename without extension
  filename=$(basename "$file".txt)
  
  # Create output file path
  output_file="$OUTPUT_DIR/$filename.json"
  
  # Run conversion script
 ./convert_prompt.sh "$file" "$output_file"
done

Word of the Day: Illusion. Art is an illusion.

Here are the translations for the word "illusion" and the sample sentence "Art is an illusion" in the requested languages:

1. French:

   • Illusion: illusion (no change)

   • Sample sentence: L'art est une illusion.

2. Spanish:

   • Illusion: ilusión

   • Sample sentence: El arte es una ilusión.

3. German:

   • Illusion: Illusion

   • Sample sentence: Die Kunst ist eine Illusion.

4. Estonian:

   • Illusion: illusioon

   • Sample sentence: Kunst on illusioon.

5. Ukrainian:

   • Illusion: ілюзія (iluziya)

   • Sample sentence: Мистецтво - це ілюзія (Mystetstvo - tse iluziya).

6. Russian:

   • Illusion: иллюзия (illyuziya)

   • Sample sentence: Искусство - это иллюзия (Iskusstvo - eto illyuziya).

7. Chinese (Simplified):

• Illusion: (huàn xiǎng)

• Sample sentence: (yì shù shì yī gè huàn xiǎng)

艺术是一个幻想。

Note: For Chinese, the transliteration is in Pinyin, which is the standard romanization system for Mandarin Chinese. Ee shoo shee ee guh hwahn hsiang.

Transliteration for non-Latin scripts:

• Ukrainian: ілюзія (iluziya)

• Russian: иллюзия (illyuziya)

• Chinese: (huàn xiǎng)