diff --git a/.github/readme.md b/.github/readme.md index 885f25461..3fa3826b4 100644 --- a/.github/readme.md +++ b/.github/readme.md @@ -223,7 +223,7 @@ If the ST-hosting device is on the same wifi network, you will use the ST-host's If you (or someone else) want to connect to your hosted ST while not being on the same network, you will need the public IP of your ST-hosting device. -* While using the ST-hosting device, access [this page](https://whatismyipaddress.com/) and look for for `IPv4`. This is what you would use to connect from the remote device. +* While using the ST-hosting device, access [this page](https://whatismyipaddress.com/) and look for `IPv4`. This is what you would use to connect from the remote device. ### 3. Connect the remote device to the ST host machine diff --git a/colab/GPU.ipynb b/colab/GPU.ipynb index 630ea43ca..4adf35e6c 100644 --- a/colab/GPU.ipynb +++ b/colab/GPU.ipynb @@ -40,20 +40,20 @@ "use_sd_cpu = False #@param {type:\"boolean\"}\n", "#@markdown ***\n", "#@markdown Loads the image captioning module\n", - "extras_enable_captioning = True #@param {type:\"boolean\"}\n", - "Captions_Model = \"Salesforce/blip-image-captioning-large\" #@param [ \"Salesforce/blip-image-captioning-large\", \"Salesforce/blip-image-captioning-base\" ]\n", + "extras_enable_caption = True #@param {type:\"boolean\"}\n", + "captioning_model = \"Salesforce/blip-image-captioning-large\" #@param [ \"Salesforce/blip-image-captioning-large\", \"Salesforce/blip-image-captioning-base\" ]\n", "#@markdown * Salesforce/blip-image-captioning-large - good base model\n", "#@markdown * Salesforce/blip-image-captioning-base - slightly faster but less accurate\n", "#@markdown ***\n", "#@markdown Loads the sentiment classification model\n", - "extras_enable_emotions = True #@param {type:\"boolean\"}\n", - "Emotions_Model = \"nateraw/bert-base-uncased-emotion\" #@param [\"nateraw/bert-base-uncased-emotion\", \"joeddav/distilbert-base-uncased-go-emotions-student\"]\n", + "extras_enable_classify = True #@param {type:\"boolean\"}\n", + "classification_model = \"nateraw/bert-base-uncased-emotion\" #@param [\"nateraw/bert-base-uncased-emotion\", \"joeddav/distilbert-base-uncased-go-emotions-student\"]\n", "#@markdown * nateraw/bert-base-uncased-emotion = 6 supported emotions
\n", "#@markdown * joeddav/distilbert-base-uncased-go-emotions-student = 28 supported emotions\n", "#@markdown ***\n", "#@markdown Loads the story summarization module\n", - "extras_enable_memory = True #@param {type:\"boolean\"}\n", - "Memory_Model = \"slauw87/bart_summarisation\" #@param [ \"slauw87/bart_summarisation\", \"Qiliang/bart-large-cnn-samsum-ChatGPT_v3\", \"Qiliang/bart-large-cnn-samsum-ElectrifAi_v10\", \"distilbart-xsum-12-3\" ]\n", + "extras_enable_summarize = True #@param {type:\"boolean\"}\n", + "summarization_model = \"slauw87/bart_summarisation\" #@param [ \"slauw87/bart_summarisation\", \"Qiliang/bart-large-cnn-samsum-ChatGPT_v3\", \"Qiliang/bart-large-cnn-samsum-ElectrifAi_v10\", \"distilbart-xsum-12-3\" ]\n", "#@markdown * slauw87/bart_summarisation - general purpose summarization model\n", "#@markdown * Qiliang/bart-large-cnn-samsum-ChatGPT_v3 - summarization model optimized for chats\n", "#@markdown * Qiliang/bart-large-cnn-samsum-ElectrifAi_v10 - nice results so far, but still being evaluated\n", @@ -63,10 +63,18 @@ "extras_enable_silero_tts = True #@param {type:\"boolean\"}\n", "#@markdown Enables Microsoft Edge text-to-speech module\n", "extras_enable_edge_tts = True #@param {type:\"boolean\"}\n", + "#@markdown Enables RVC module\n", + "extras_enable_rvc = False #@param {type:\"boolean\"}\n", + "#@markdown ***\n", + "#@markdown Enables Whisper speech recognition module\n", + "extras_enable_whisper_stt = True #@param {type:\"boolean\"}\n", + "whisper_model = \"base.en\" #@param [ \"tiny.en\", \"base.en\", \"small.en\", \"medium.en\", \"tiny\", \"base\", \"small\", \"medium\", \"large\" ]\n", + "#@markdown There are five model sizes, four with English-only versions, offering speed and accuracy tradeoffs.\n", + "#@markdown The .en models for English-only applications tend to perform better, especially for the tiny.en and base.en models.\n", "#@markdown ***\n", "#@markdown Enables SD picture generation\n", "extras_enable_sd = True #@param {type:\"boolean\"}\n", - "SD_Model = \"ckpt/anything-v4.5-vae-swapped\" #@param [ \"ckpt/anything-v4.5-vae-swapped\", \"hakurei/waifu-diffusion\", \"philz1337/clarity\", \"prompthero/openjourney\", \"ckpt/sd15\", \"stabilityai/stable-diffusion-2-1-base\" ]\n", + "sd_model = \"ckpt/anything-v4.5-vae-swapped\" #@param [ \"ckpt/anything-v4.5-vae-swapped\", \"hakurei/waifu-diffusion\", \"philz1337/clarity\", \"prompthero/openjourney\", \"ckpt/sd15\", \"stabilityai/stable-diffusion-2-1-base\" ]\n", "#@markdown * ckpt/anything-v4.5-vae-swapped - anime style model\n", "#@markdown * hakurei/waifu-diffusion - anime style model\n", "#@markdown * philz1337/clarity - realistic style model\n", @@ -91,28 +99,36 @@ "if secure:\n", " params.append('--secure')\n", "params.append('--share')\n", - "ExtrasModules = []\n", + "modules = []\n", "\n", - "if (extras_enable_captioning):\n", - " ExtrasModules.append('caption')\n", - "if (extras_enable_memory):\n", - " ExtrasModules.append('summarize')\n", - "if (extras_enable_emotions):\n", - " ExtrasModules.append('classify')\n", - "if (extras_enable_sd):\n", - " ExtrasModules.append('sd')\n", - "if (extras_enable_silero_tts):\n", - " ExtrasModules.append('silero-tts')\n", + "if extras_enable_caption:\n", + " modules.append('caption')\n", + "if extras_enable_summarize:\n", + " modules.append('summarize')\n", + "if extras_enable_classify:\n", + " modules.append('classify')\n", + "if extras_enable_sd:\n", + " modules.append('sd')\n", + "if extras_enable_silero_tts:\n", + " modules.append('silero-tts')\n", "if extras_enable_edge_tts:\n", - " ExtrasModules.append('edge-tts')\n", - "if (extras_enable_chromadb):\n", - " ExtrasModules.append('chromadb')\n", + " modules.append('edge-tts')\n", + "if extras_enable_chromadb:\n", + " modules.append('chromadb')\n", + "if extras_enable_whisper_stt:\n", + " modules.append('whisper-stt')\n", + " params.append(f'--stt-whisper-model-path={whisper_model}')\n", + "if extras_enable_rvc:\n", + " modules.append('rvc')\n", + " params.append('--max-content-length=2000')\n", + " params.append('--rvc-save-file')\n", "\n", - "params.append(f'--classification-model={Emotions_Model}')\n", - "params.append(f'--summarization-model={Memory_Model}')\n", - "params.append(f'--captioning-model={Captions_Model}')\n", - "params.append(f'--sd-model={SD_Model}')\n", - "params.append(f'--enable-modules={\",\".join(ExtrasModules)}')\n", + "\n", + "params.append(f'--classification-model={classification_model}')\n", + "params.append(f'--summarization-model={summarization_model}')\n", + "params.append(f'--captioning-model={captioning_model}')\n", + "params.append(f'--sd-model={sd_model}')\n", + "params.append(f'--enable-modules={\",\".join(modules)}')\n", "\n", "\n", "%cd /\n", @@ -121,22 +137,13 @@ "!git clone https://github.com/Cohee1207/tts_samples\n", "!npm install -g localtunnel\n", "%pip install -r requirements.txt\n", - "#!pip install tensorflow==2.14\n", - "#!pip install colorama\n", - "#!pip install Flask-Cors\n", - "#!pip install Flask-Compress\n", - "#!pip install transformers\n", - "#!pip install Flask_Cloudflared\n", - "#!pip install webuiapi\n", - "#!pip install diffusers\n", - "#!pip install accelerate\n", - "#!pip install silero_api_server\n", - "#!pip install edge_tts\n", - "#!pip install chromadb\n", - "#!pip install sentence_transformers\n", "!wget https://github.com/cloudflare/cloudflared/releases/download/2023.5.0/cloudflared-linux-amd64 -O /tmp/cloudflared-linux-amd64\n", "!chmod +x /tmp/cloudflared-linux-amd64\n", "\n", + "if extras_enable_rvc:\n", + " print(\"Installing RVC requirements\")\n", + " !pip install -r requirements-rvc.txt\n", + "\n", "# Generate a random API key\n", "api_key = secrets.token_hex(5)\n", "\n", diff --git a/public/index.html b/public/index.html index 6ac83f389..78c1dfdf8 100644 --- a/public/index.html +++ b/public/index.html @@ -8,6 +8,7 @@ + diff --git a/public/manifest.json b/public/manifest.json new file mode 100644 index 000000000..28df3de4c --- /dev/null +++ b/public/manifest.json @@ -0,0 +1,30 @@ +{ + "name": "SillyTavern", + "short_name": "SillyTavern", + "start_url": "/", + "display": "standalone", + "theme_color": "#202124", + "background_color": "#202124", + "icons": [ + { + "src": "img/apple-icon-57x57.png", + "sizes": "57x57", + "type": "image/png" + }, + { + "src": "img/apple-icon-72x72.png", + "sizes": "72x72", + "type": "image/png" + }, + { + "src": "img/apple-icon-114x114.png", + "sizes": "114x114", + "type": "image/png" + }, + { + "src": "img/apple-icon-144x144.png", + "sizes": "144x144", + "type": "image/png" + } + ] +} diff --git a/public/scripts/extensions/shared.js b/public/scripts/extensions/shared.js index 37dbec96a..26164d8d0 100644 --- a/public/scripts/extensions/shared.js +++ b/public/scripts/extensions/shared.js @@ -23,7 +23,7 @@ export async function getMultimodalCaption(base64Img, prompt) { const compressionLimit = 2 * 1024 * 1024; if (extension_settings.caption.multimodal_api === 'openrouter' && base64Bytes > compressionLimit) { const maxSide = 1024; - base64Img = await createThumbnail(base64Img, maxSide, maxSide); + base64Img = await createThumbnail(base64Img, maxSide, maxSide, 'image/jpeg'); } const apiResult = await fetch('/api/openai/caption-image', { diff --git a/public/scripts/utils.js b/public/scripts/utils.js index d8b99956f..571583240 100644 --- a/public/scripts/utils.js +++ b/public/scripts/utils.js @@ -973,9 +973,10 @@ export function loadFileToDocument(url, type) { * @param {string} dataUrl The data URL encoded data of the image. * @param {number} maxWidth The maximum width of the thumbnail. * @param {number} maxHeight The maximum height of the thumbnail. + * @param {string} [type='image/jpeg'] The type of the thumbnail. * @returns {Promise} A promise that resolves to the thumbnail data URL. */ -export function createThumbnail(dataUrl, maxWidth, maxHeight) { +export function createThumbnail(dataUrl, maxWidth, maxHeight, type = 'image/jpeg') { return new Promise((resolve, reject) => { const img = new Image(); img.src = dataUrl; @@ -1000,7 +1001,7 @@ export function createThumbnail(dataUrl, maxWidth, maxHeight) { ctx.drawImage(img, 0, 0, thumbnailWidth, thumbnailHeight); // Convert the canvas to a data URL and resolve the promise - const thumbnailDataUrl = canvas.toDataURL('image/jpeg'); + const thumbnailDataUrl = canvas.toDataURL(type); resolve(thumbnailDataUrl); };