Otter
squeezelite-esp32
Otter | squeezelite-esp32 | |
---|---|---|
4 | 8 | |
3,454 | 963 | |
- | - | |
9.1 | 9.2 | |
2 months ago | about 1 month ago | |
Python | C | |
MIT License | - |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
Otter
-
OpenAI vs Google, Detect ChatGPT Content with 99% accuracy, Navigating AI compute costs
๐ Video-LLaMA - Empower large language models with video and audio understanding capability. (link) ๐ฆฆ Otter - Multi-modal model with improved instruction-following and in-context learning ability. ๐ Linkly.AI - AI-powered lead analytics and management platform that helps you track, analyze, and streamline your leads in one place. ๐ฌ Jet Cut Ready - AI plugin for Adobe Premiere Pro that automatically removes silent parts in videos. (link) ๐ฌ HeyGen's ChatGPT Plugin - Convert text into high-quality videos using AI text and video generation.
- Multimodal models and "active" learning
- Otter: A Multi-Modal Model with In-Context Instruction Tuning
-
Otter is a multi-modal model developed on OpenFlamingo (open-sourced version of DeepMind's Flamingo), trained on a dataset of multi-modal instruction-response pairs. Otter demonstrates remarkable proficiency in multi-modal perception, reasoning, and in-context learning.
GitHub repo includes HuggingFace links to the model: https://github.com/Luodian/Otter
squeezelite-esp32
-
[Advice request] DIY Airplay / Bluetooth receiver with RPi or ESP32 or ?
I built an Airplay receiver with RPi (+SPDIF out) using Volumio but it takes for ever to boot and often does not work, the mpd daemon crashes somehow (it reindex all the audio files on boot) I read about https://github.com/sle118/squeezelite-esp32 but I don't know about boot times.
- Cool embedded + music/audio related projects you'd like to share?
-
Logitech Media Server 8.3.0 has received a new beta release dated 31 October
and apparently an esp32! https://github.com/sle118/squeezelite-esp32
-
esp32 flashing with squeezelite-esp32
...and so on. I've completely erased the device before attempting to write flash memory, and have confirmed that it's 100% written. Have tried at low baud rate, with multiple micro-USB cables, and have tried using both windows PC and my macbook, all with the same results. I've also tried the whole gamut of possible firmwares from the repo. First board was a TTGO T8 V1.8 ESP32 board, second one a generic ESP32-WRover Dev board from amazon. Any tips on flashing with this firmware? Or any recommendations of a wrover dev board that may give me less trouble?
-
esp32-audio-kit
I got one of the AI thinker esp32-A1S audio kits (https://docs.ai-thinker.com/en/esp32-audio-kit) which I was hoping to run as a DAC for a logitech media server using squeezelite-esp32. Apparently these boards are quite finicky and a bit challenging to flash, but having tried all the obvious fixes and various ideas that duckduckgo had to offer, I'm still stuck in a boot loop. I've got dedicated power going into USB1 and connected to a USB-C port on my MacBook via USB2. I've tried with the dip switches all up, all down, and in the default 3 up / 2 down configuration, each time erasing the flash memory completely before trying to load the squeezelite-esp32 binary. Using ESPHome-Flasher, I get the following dialogue:
-
Yet another DT770 internal Bluetooth conversion!
In theory yes, but it is more of a 'whole summer project' than 'one evening project'. There are small ESP32 Wi-Fi modules that could be connected to small I2S headphones amplifier. Both should fit inside bigger cups like DT770 has if you remove acoustic padding. You also would have to build custom dock for mixing audio inputs, converting them to digital and pushing data through Wi-FI, possibly also ESP32 based with external ADCs. The hardest part will be writing custom code to glue it all together. There are some ESP32 Wi-Fi streaming libs like: https://github.com/sle118/squeezelite-esp32 This would be a good starting point, but the amount of customisation to achieve what you described will be non-trivial. Battery life of headphones would not be stellar either, ESP modules are more suited to be used in smart home appliances powered from the grid. I'd guess about 3 hours with 750mAh battery while streaming over Wi-Fi.
-
Let's make a definitive guide to the subtle differences in Self Hosted Music Streaming.
Squeezelite has also been ported to the ESP32, so one could probably create (or recreate) the old Logitech/Slim Devices hardware clients, VFD displays and all.
-
Speaker similar to Sonos One but without the... Sonos
Here's what I'm looking for: * Powered speaker of similar form factor/aesthetic as the (white) Sonos One. Must look nice (aesthetic is as important as sound quality). * Ideally would have an amp per speaker. Most bookshelf's have a single amp in one speaker with amplified sound run via speaker wire to the other. Would prefer to not have this. But, worst case, I could buy a couple sets and use the powered speakers where I can't run speaker wire, and have a pair of passive speakers for a room where I can. Not the end of the world. * Speakers that are better for sourceless, far-field listening. Most bookshelfs in this vein seem tailored to near field listening. * Have digital input so I can go cheap/small on the squeezeplayer setup (would love to use the ESP32 squeezelite) without worrying about audio quality (either DAC or amplification).
What are some alternatives?
LLaMA-Adapter - [ICLR 2024] Fine-tuning LLaMA to follow Instructions within 1 Hour and 1.2M Parameters
mpd - Music Player Daemon
NExT-GPT - Code and models for NExT-GPT: Any-to-Any Multimodal Large Language Model
Jellyfin - The Free Software Media System
Video-LLaMA - [EMNLP 2023 Demo] Video-LLaMA: An Instruction-tuned Audio-Visual Language Model for Video Understanding
LMS - Lightweight Music Server. Access your self-hosted music using a web interface.
Sophia - Effortless plugin and play Optimizer to cut model training costs by 50%. New optimizer that is 2x faster than Adam on LLMs.
Navidrome Music Server - ๐งโ๏ธ Modern Music Server and Streamer compatible with Subsonic/Airsonic
Awesome-Multimodal-Large-Language-Models - :sparkles::sparkles:Latest Papers and Datasets on Multimodal Large Language Models, and Their Evaluation.
gonic - music streaming server / free-software subsonic server API implementation
LinkedInGPT - Skynet
LMS - LMS written in C++17