@@greatscottlab here's a really good video idea: HOW TO MAKE VCO SINE. tiny arduinos aren't fast enough. circuits that support variable pitch are complex. i am not an electronics expert but i am a sound expert. i want to make a sine. i don't want to spend $30 in parts ot make a sine. i want a sine. give me a sine. someone make a video on how to make variable pitch sine oscillator with few and inexpensive parts. i don't want to include an incandescant bulb in the circuit. i just want a sine oscillator i can change the pitch of across the audio range. i will measure it with an arduino. i don't want a sine with a static pitch. i don't want to control it with MIDI. i just want to plop a potentiometer, on a circuit, and output a sine wave of variable frequency. how is this so complicated? i have 20+ years of audio dsp. i know a T bridge circuit will produce a sine because its an elementary structure parallelled in dsp. where th eufck are the fucking cheap to build sine oscillators? what the fucking fuck? why the fuck can;t people understand that you would want an oscillator with a variable pitch and a sine contour/spectrum? how does this not happen in electronics, which is just physics? i don't want three phase power. i don't want a 741 op amp. i want cheap ass parts, as few as possible. i want to do this shit. i'm fucking fifty i should have done it thirty years ago. fucking hell. it's just one frequency.
@@atomictraveller Did you get anywhere with your sine generator? Basically, you need a couple of tables of constants, which could be filled in by calculations when it boots up, and a variable slope ramp / saw wave generator, which is used as an index into the sine table. 1) Calculate a step value, based on the ADC value from the pot, maybe using a lookup table to speed things up, if you want something like V/octave scaling. 2) In your main loop, keep adding that step to your index value, and loop it when it reaches the maximum. 3) Use a DAC of some sort, or I2S, to output the value, from the table. As a nice side effect, you also get a VCO saw output, if you have a second DAC - or use the other channel of I2S. Due to obvious symmetry, you don't necessarily need a full period of a sine wave. A quarter wave will do. You might need to disable interrupts, to avoid jitter caused by the CPU being too busy.
@@TooSlowTube i generally render sines in dsp with quadrature... s0 = 1; s1 = 0; // init s0 -= w * s1; s1 += w * s0; // loop where w (angular freq) = 2*pi*hz/samplerate
my pic32 had enough flash to store a funny audio sample and i managed to port bitluni's arduino audio sampling code to it. Bitluni and Greatscott are great teachers !
Fun fact: You can use I2S to generate the 38 kHz carrier for IR signaling, and save a lot of CPU cycles compared to the popular techniques. (One of the features under the hood is called DMA, I think it would be worth a separate video)
Awesome :) . Planning to add darth vader sounds to my room Automation project.. This will be really helpful.... The force is strong with Scott !!! Love from India
I keep thinking about re-creating the Portal Radio (from the game) and something like this would work a treat (minus the microphone). Only want it to play the standard low-quality tune on loop. Great video.
well there are also decoder chips that play mp3's www.adafruit.com/product/1788 this is a good board to work with it also has theoretically midi syntesizer capabilities and the library has all the features needed. works with an arduino which for me is an added bonus.
Thanks for responding to my request Greatscott. I actually hold a bachelors in Electronics but couldn't get into that field due to some reasons. I'm a datascientist now and I'm able to still have a good hold on Electronics is just because of your videos. Stay creative, stay hungry for more knowledge
It's important to note that while greater sampling frequency does increase quality, any sampling frequency above 40 kHz makes irrelevant increase to the human hearing. Because, according do the Nyquist theorem, to perfectly reconstruct a analog signal you need a sampling frequency at least double the analysed signal. Thus we need only 40kHz, as the human hearing only goes to about 20kHz (and that's in very young children, because you'd be lucky to have this hearing past your adolescence). Also speakers are almost always limited to playing till this frequency, so even if you say that frequencies above 20kHz do somehow contribute to sound quality, your speakers won't even play it.
@Heads MessRegarding your point about two points making a triangle/sawtooth wave: Whole point of the nyquist theorem is that IF we are certain that no frequencies greater than the nyquist frequency was contained within the signal before we sample it, then we can reconstruct the sampled signal with any temporal resolution we want. So you dont have to drive your speaker with those 2 points, since nyquist tells you that thre is no ambiguity about the frequencies in there, you can fill the samples in between to form the sinusoidal before you drive the speaker. And actually you dont even have to bother about doing that in software as a lowpass filter on the sampled output will do that for you anyways. I agree with the rest of your comment.
Hmmm I guess it could be done... but bluetooth I think it's limited to about 10m range, and wi-fi would require one device to act as an AP and others to trust it, also said device would have to be within a reasonable distance of every other device... seems like an overkill for something that could be done just over radio
How convenient it was to find this nice video literally 24 hours after I figured out how to code I2S on ESP32 and how to connect wires to external DAC. LOL. Oh man... I'd say wiring it together was the easiest part. The code and RTOS required a bit of wrestling...
@@drescherjm and me personally I get very interested in knowing what the problem is, sometimes it can give more insight when we are doing our own projects
@@ayushp.5395 lol obviously a waste. Who spends so much for headphones? You need high quality audio files like FLAC for these headphones to work in full potential... Waste of money. My 3€ earbuds works perfectly.
Hi, I don't understand why the sound is so bad at the end of the video as you used 16bits /44khz which should result with good audio quality? Is it a limitation on the esp32? Is it because of your speaker?
Mostly likely not using DMA with double buffering. Without that any delay will show up as breaks. In realtime transmission, it's a must. If playing from a file, still the IO will cause lots of interrupts and result in breaking audio, even with a fast mcu.
Hey GreatScott I just recently created a HiFi bluetooth receiver with i2s and a Raspberry with a Hifiberry DAC+ DSP hat. The reason to use this method was the quality and input methods. The CSRA6415 I used can handle APTX which is a high quality bluetooth protocol. Also this chip can handle button Inputs (to play / pause or control volume on the smartphone via local buttons) as well as many other things. It's a very cheap but great way! It's completely configurable. So I connected it to my Hifiberry hat with correct I2S parameters and it worked a treat. Maybe an upgrade for your portable speaker! :) I just realized I could have written this in Deutsch.
@@steameier8590 What are you interested in? For information about the CSRA6415 I would recommend the suredspratz site if you search on Google. If you need information about the tool, there's none. It's trial and error but theres a great standard configuration preloaded. For the Raspberry I used Moode as an OS and the hifiberry DAC+ DSP Hat, which would be programmable but I decided to just adapt the CSRA instead of struggling with this too. Just tell me about what you would like to know more. I also just did basic Internet research and a lot of testing! :)
WOW WOW WOW I just checked the price of your Oscilloscope MSOX3054T (500Mhz). It's about 15,000USD! So expensive! Please advice me a cheap and economic one. I need it for learning as you are teaching us in your nice videos. Also I need it for repairing home audio systems. Thanks GreatScott! Stay creative!
Interesting, a month ago I set up a streaming player with an old raspberry Pi and a TeraDaK ES9023. The Pi is an old model B, 26 pin header but has an unpopulated 8 pin header for I2S. Sounds great and was pretty easy to set up.
Great video. I really like this project, just what I was looking for. I just wish you had a project page that showed how to build it like the parts and code used.
Wie geht herr Scott! Man, you got some pretty informative stuff going on but I think about 70 percent off it is going over our laymans heads! Maybe the electronics majors here have an easy time understanding your tech terms but as easily as you explain the rest of us probably just pretend we understood. Just saying you should make some beginner or intro into electronics videos and your viewership will grow! Danke fur die lektionen!
I’m disappointed. From an implementation standpoint, this leaves more questions than answers. Great if you use an ESP32 and that particular audio library, but this doesn’t really explain I2S in a way which allows competent viewers to use I2S peripherals with generic microcontrollers.
I want to add a little to the whole sampling rate debate. 44.1khz is all you need. It is in fact enough to PERFECTLY represent an audio signal ranging between 0Hz and 20khz. You do not need more. It is completely wasted to add more. Reason: you only need twice as many sampling points as the highest frequency you want to digitise. The rest can be mathematically interpolated. Higher Bit Depth can be usefull for higher volumes. Also it gives more freedom when mastering the song. However, as a listener you only need a certain amount of dynamic range. 16Bit is completely fine. You may argue for 24Bit but anything above is only useful at volumes that would completely destroy your ears.
This might work well for my Halloween decoration I created years ago. I use a raspberry pi to play audio and flash red LEDs in the eyes when a motion event happens from my security camera. I've always felt like the raspberry pi was overkill. All it really does is has a webserver for commands. Takes the commands and plays either a random file (for motion) or a specific file from the command. Analyzes the wave form to blink the LEDs in sync with what is being said. The only thing that might be tricky is I have a web interface set up and I can talk through it using the microphone on my mobile phone. But it seems like I can probably get rid of the raspberry pi and just use an ESP32. Get rid of the hosted webserver and use my existing MQTT.
Why was the output sound so disrupted? I was expecting that i2s is pretty much lossless. I am thinking about creating wireless 5.1 sound system (it will use only power cords) using my old 5.1 (with too much cables) and using esp32 and esp-now. Problem is that I need to convert 5.1 audio to i2s and then distribute that to each speaker. Any ideas? It doesn't need to be esp32, just everything which would make it wireless 5.1 and is already documented.
Your content is really great!. Please make a video on "Commercial grade Low Power IC's (Bluetooth and wifi) for low power IoT". Please compare them with their specs like current drawn while transmission and reception. I would love to get insights about the topic, from you. Thank you!
Haven't played with ESP32 boards yet. I've dome alot with the arduino and the dfplayer mini. Sound quality isn't as good, but works for the small projects.
I've got a couple esp32's kicking around but haven't used them yet... although that's gonna be changing soon since the 8266 has a bunch of unusable ports on it
The other day i,'ve tried to attach the i2s dac to my router. Finally the Linux community have fixed the dma driver and now my router can play music :D
@@greatscottlab Yes, my router is based around the mt7628an chip. It have pcie x1, lots of gpios, i2s or pcm. It also have 2r2t wifi 2.4Ghz. If you run Openwrt you can do a lot of cool things!
Could you use the two 8 bit DACs and combine them with the R-2R resistor arrangement to make a higher resolution output? A downside would be only 1 channel, no stereo
Good afternoon Mr. Scott, you went all the way to be able to playback a 16bit, 44.1kHz, stereo sound and the quality is appalling like that? That is surely no better than a mono 8bit output. In fact it is reminiscent of audio output of the Commodore 64 when the audio chip was turned on and off quickly enough to transform pulses in actual speech. I have worked with I2S equipped boards that have perfectly good audio quality, that one is unacceptable - for the reasons explained in the several first minutes of the video. I have reasons to think if you used a lower sampling rate and resolution you'd get a better result.
Also have you written the Device Driver for PCM5122 ? How are you providing the I2S clocks ? What is your CLK source ? Can you support 32-bit PCM instead of 16-bit PCM..?
Finally a crystal clear explanation of I2S!
You're welcome :-)
@@greatscottlab here's a really good video idea: HOW TO MAKE VCO SINE. tiny arduinos aren't fast enough. circuits that support variable pitch are complex. i am not an electronics expert but i am a sound expert. i want to make a sine. i don't want to spend $30 in parts ot make a sine. i want a sine. give me a sine. someone make a video on how to make variable pitch sine oscillator with few and inexpensive parts. i don't want to include an incandescant bulb in the circuit. i just want a sine oscillator i can change the pitch of across the audio range. i will measure it with an arduino. i don't want a sine with a static pitch. i don't want to control it with MIDI. i just want to plop a potentiometer, on a circuit, and output a sine wave of variable frequency. how is this so complicated? i have 20+ years of audio dsp. i know a T bridge circuit will produce a sine because its an elementary structure parallelled in dsp. where th eufck are the fucking cheap to build sine oscillators? what the fucking fuck? why the fuck can;t people understand that you would want an oscillator with a variable pitch and a sine contour/spectrum? how does this not happen in electronics, which is just physics? i don't want three phase power. i don't want a 741 op amp. i want cheap ass parts, as few as possible. i want to do this shit. i'm fucking fifty i should have done it thirty years ago. fucking hell. it's just one frequency.
@@atomictraveller Did you get anywhere with your sine generator?
Basically, you need a couple of tables of constants, which could be filled in by calculations when it boots up, and a variable slope ramp / saw wave generator, which is used as an index into the sine table.
1) Calculate a step value, based on the ADC value from the pot, maybe using a lookup table to speed things up, if you want something like V/octave scaling.
2) In your main loop, keep adding that step to your index value, and loop it when it reaches the maximum.
3) Use a DAC of some sort, or I2S, to output the value, from the table.
As a nice side effect, you also get a VCO saw output, if you have a second DAC - or use the other channel of I2S.
Due to obvious symmetry, you don't necessarily need a full period of a sine wave. A quarter wave will do.
You might need to disable interrupts, to avoid jitter caused by the CPU being too busy.
@@TooSlowTube i generally render sines in dsp with quadrature...
s0 = 1; s1 = 0; // init
s0 -= w * s1; s1 += w * s0; // loop
where w (angular freq) = 2*pi*hz/samplerate
@@atomictraveller Cool. Are you using dedicated DSP hardware, or just something like a fairly fast microcontroller with an FPU?
6:38 Looks like you interpreted the signed 16 bit value returned by the mic as an unsigned one.
Maybe that would explain the clipping in his result. I was looking through comments to see an explaination. This could be it.
You are right.
9:33 The result sounds like a Dalek..
Exterminate !!
Plot twist: The Daleks also watched this video and made a sound device according to this video. That's why they sound like this.
This is just the beginning.
sounds like bsod
Touch this time text every 1 sec.
I'm so glad you finally dive into the world of ESPs!! And for a first test I2S is really not the easiest project! Thanks a lot 🙏🏻
Aight, we found the time traveler.
@@Derpuwolf Patreon.
@@richardlighthouse5328 Oh really???
This is far from his first dive into ESP
It's
He littelary connected 3 bought parts and used existing library's basic example
It's not harder then connecting addressable leds
As a high school student who only recently got into electronics, if even I can follow along with what he's saying then this is truly high quality.
Man! Those drawings are amazing. Worth watching just to see how neat the guy draws.
my pic32 had enough flash to store a funny audio sample and i managed to port bitluni's arduino audio sampling code to it. Bitluni and Greatscott are great teachers !
great stuff. would love to see more videos using the esp32!
Thanks! Will do!
Fun fact: You can use I2S to generate the 38 kHz carrier for IR signaling, and save a lot of CPU cycles compared to the popular techniques.
(One of the features under the hood is called DMA, I think it would be worth a separate video)
The esp32 has a remote control peripheral.
Awesome :) . Planning to add darth vader sounds to my room Automation project.. This will be really helpful.... The force is strong with Scott !!! Love from India
Sounds great!
I keep thinking about re-creating the Portal Radio (from the game) and something like this would work a treat (minus the microphone). Only want it to play the standard low-quality tune on loop. Great video.
well there are also decoder chips that play mp3's
www.adafruit.com/product/1788 this is a good board to work with it also has theoretically midi syntesizer capabilities and the library has all the features needed. works with an arduino which for me is an added bonus.
In pandemic times, a clip of The Police's Don't Stand So Close to Me triggered by a hidden momentary switch would be fun.
@@Ramog1000 there should be cheaper DF robot boards that outputs a headphone signal
Marvin W - Depends if you’re a Patreon? If not, sign up now!
Janus Kobain - I’m making a note here “huge success”...
Thanks for responding to my request Greatscott.
I actually hold a bachelors in Electronics but couldn't get into that field due to some reasons.
I'm a datascientist now and I'm able to still have a good hold on Electronics is just because of your videos.
Stay creative, stay hungry for more knowledge
It's important to note that while greater sampling frequency does increase quality, any sampling frequency above 40 kHz makes irrelevant increase to the human hearing. Because, according do the Nyquist theorem, to perfectly reconstruct a analog signal you need a sampling frequency at least double the analysed signal. Thus we need only 40kHz, as the human hearing only goes to about 20kHz (and that's in very young children, because you'd be lucky to have this hearing past your adolescence). Also speakers are almost always limited to playing till this frequency, so even if you say that frequencies above 20kHz do somehow contribute to sound quality, your speakers won't even play it.
@Heads MessRegarding your point about two points making a triangle/sawtooth wave: Whole point of the nyquist theorem is that IF we are certain that no frequencies greater than the nyquist frequency was contained within the signal before we sample it, then we can reconstruct the sampled signal with any temporal resolution we want. So you dont have to drive your speaker with those 2 points, since nyquist tells you that thre is no ambiguity about the frequencies in there, you can fill the samples in between to form the sinusoidal before you drive the speaker. And actually you dont even have to bother about doing that in software as a lowpass filter on the sampled output will do that for you anyways.
I agree with the rest of your comment.
Can you make a walkie talkie with 2 esp32's using this metod and wifi/bluetooth?
I can put it on my to do list
@@greatscottlab I'd love to see that.
Yes!
Ooo. Hack up some headsets perhaps?
Hmmm I guess it could be done... but bluetooth I think it's limited to about 10m range, and wi-fi would require one device to act as an AP and others to trust it, also said device would have to be within a reasonable distance of every other device... seems like an overkill for something that could be done just over radio
How convenient it was to find this nice video literally 24 hours after I figured out how to code I2S on ESP32 and how to connect wires to external DAC. LOL. Oh man... I'd say wiring it together was the easiest part. The code and RTOS required a bit of wrestling...
9:32 Save yourselves, Great Scott has already been terminated 😂.
Yaaaaaaaaaaaaaaaay new awesome video i love your Videos they are beeeeeeeest
Helpful video 👍
Do you know why it sounded so weird in the end? Great video, btw!
I agree from the video I expected it to sound much better.
@@drescherjm and me personally I get very interested in knowing what the problem is, sometimes it can give more insight when we are doing our own projects
Sounds like jitter
Probably either unwanted interrupts or reading from sdcard took too long
We need a revisit when he figures it out. This is not acceptable quality from i2s
I love your videos nd loved this one too.
Had been a subscriber of yours since 3 years
Awesome :-)
6:48 whoopsi , someone soldered on his probe :D
You know i was waiting for this one 🤩
Wow! This video is really awesome. Didn't know the ESP had such capabilities.
I learn so much from your videos
I love these kind of projects great scott keep it up😍😍😍😍😍🔥🔥🔥🔥
Glad you like them!
1:26 Was the pun intended?
Maybe
well, so my $1200 24bit/192KHz DAC and $990 wired headphones were a waste of money???
not to forget the $20/month subscription for tidal...
@@ayushp.5395 lol obviously a waste. Who spends so much for headphones? You need high quality audio files like FLAC for these headphones to work in full potential...
Waste of money. My 3€ earbuds works perfectly.
EdvisKing Come on! 64 bit sound all the way!
I was just about to do this with these exact materials, perfect timing
Great video, and the best I2C explanation , thank you for sharing
Hi, I don't understand why the sound is so bad at the end of the video as you used 16bits /44khz which should result with good audio quality? Is it a limitation on the esp32? Is it because of your speaker?
Mostly likely not using DMA with double buffering. Without that any delay will show up as breaks. In realtime transmission, it's a must. If playing from a file, still the IO will cause lots of interrupts and result in breaking audio, even with a fast mcu.
I'm not sure, but I think the is2 amplifier generates PWM output instead of real waveform.
Hey GreatScott
I just recently created a HiFi bluetooth receiver with i2s and a Raspberry with a Hifiberry DAC+ DSP hat.
The reason to use this method was the quality and input methods.
The CSRA6415 I used can handle APTX which is a high quality bluetooth protocol. Also this chip can handle button Inputs (to play / pause or control volume on the smartphone via local buttons) as well as many other things.
It's a very cheap but great way! It's completely configurable.
So I connected it to my Hifiberry hat with correct I2S parameters and it worked a treat.
Maybe an upgrade for your portable speaker! :)
I just realized I could have written this in Deutsch.
Do you have some links to this topics?
@@steameier8590 What are you interested in? For information about the CSRA6415 I would recommend the suredspratz site if you search on Google. If you need information about the tool, there's none. It's trial and error but theres a great standard configuration preloaded. For the Raspberry I used Moode as an OS and the hifiberry DAC+ DSP Hat, which would be programmable but I decided to just adapt the CSRA instead of struggling with this too.
Just tell me about what you would like to know more. I also just did basic Internet research and a lot of testing! :)
Thank you very much! We really enjoy your videos. You are badass and Nationwide!... Joel
WOW WOW WOW I just checked the price of your Oscilloscope MSOX3054T (500Mhz). It's about 15,000USD! So expensive! Please advice me a cheap and economic one. I need it for learning as you are teaching us in your nice videos. Also I need it for repairing home audio systems. Thanks GreatScott! Stay creative!
You've been hitting home runs with the projects you have been choosing. Thanks for all your hard work. The ESP32 is pretty amazing, but very complex.
Love your vids bought my first arduino thanks to you
Glad I could help
Great info if you're wondering how modern car audio works.
Great video as always!
Whoa amazing explanation dude!
I can't Guess how much paper do you use for every video 😂😂 Aniway, you are the eletronic teacher i never had at school, Simply awesome!
You are really awesome iam a great fan of you.thank very much for sharing your knowledge
It's my pleasure
Good video as always but I would have like to seen why it sounds distorted and how you get around it.
Interesting, a month ago I set up a streaming player with an old raspberry Pi and a TeraDaK ES9023. The Pi is an old model B, 26 pin header but has an unpopulated 8 pin header for I2S. Sounds great and was pretty easy to set up.
Thank you! Been wondering why folks were using this for audio, haha turns out that's what it is for! :D
Thanks gonna use this exact dac to make a synth (nothing to complicated probably two osc max). Should be fun.
You make it look oh so easy.
Sir iam an huge fan of u sir keep creating, more viedoes like this 😍
Awesome!!! If you come to Mexico someday, i'll invite you "unos tacos"
Sounds good!
Really interesting!!! Thanks a lot, dude!!! 😃
Stay safe there! 🖖😊
awesome project, Scott! Thx!
Great walkthrough 👍
Thanks for sharing 👍😀
Great video. I really like this project, just what I was looking for. I just wish you had a project page that showed how to build it like the parts and code used.
A good follow-up to this basics video: Direct Memory Access (or DMA). For very high speed data transfer
Wie geht herr Scott! Man, you got some pretty informative stuff going on but I think about 70 percent off it is going over our laymans heads! Maybe the electronics majors here have an easy time understanding your tech terms but as easily as you explain the rest of us probably just pretend we understood. Just saying you should make some beginner or intro into electronics videos and your viewership will grow! Danke fur die lektionen!
What else to say here than: Great Scott!
Thanks for the great explanation of i2s!
I liked it 👍 ,great content
Glad you enjoyed
1:44 Why not use both, decrease the level with a voltage divider and add using an OpAmp?!
youre a legend thanks for this
Great tutorial !!👍
thanks finally I can build something this weekend.
GreatScott get Datasheet
GreatScott *_HIGHLIGHT_*
great video until at the end the audio that came out was all choppy! what was the reason for that? Thanks!! can it sound better than that? :)
I’m disappointed. From an implementation standpoint, this leaves more questions than answers. Great if you use an ESP32 and that particular audio library, but this doesn’t really explain I2S in a way which allows competent viewers to use I2S peripherals with generic microcontrollers.
Very cool!
Könntest du mehr zum Thema Audio machen?
If you want to make a 16 bit resistor DAC, you would needd 65536 pins, and double the resistors. Good video
I'm sorry nobody corrected you, you need only 16 pins, you can use a SIPO shift register to bring that number down.
You let us Starve for your videos.. please upload more often
Any idea how much time it takes to produce one video with this quality?
@@greatscottlab I know it takes time but can't to do something to reduce the time gap. It will be beneficial for you as well as for viewers
No
😓
Thank you! Thank you! Thank you!
please do a video at Resistor Ladder R2R DACs for high quality music reproduction! great video
I want to add a little to the whole sampling rate debate. 44.1khz is all you need. It is in fact enough to PERFECTLY represent an audio signal ranging between 0Hz and 20khz. You do not need more. It is completely wasted to add more. Reason: you only need twice as many sampling points as the highest frequency you want to digitise. The rest can be mathematically interpolated.
Higher Bit Depth can be usefull for higher volumes. Also it gives more freedom when mastering the song. However, as a listener you only need a certain amount of dynamic range. 16Bit is completely fine. You may argue for 24Bit but anything above is only useful at volumes that would completely destroy your ears.
Can you make the quality of the audio better with some modification? i was kinda disappointed with that much work for a so "robot"-like sound
at 6:30 the microphone you have the wrong side... the hole in the centre is the opening (there is the mic logo^^)
This might work well for my Halloween decoration I created years ago. I use a raspberry pi to play audio and flash red LEDs in the eyes when a motion event happens from my security camera. I've always felt like the raspberry pi was overkill. All it really does is has a webserver for commands. Takes the commands and plays either a random file (for motion) or a specific file from the command. Analyzes the wave form to blink the LEDs in sync with what is being said. The only thing that might be tricky is I have a web interface set up and I can talk through it using the microphone on my mobile phone. But it seems like I can probably get rid of the raspberry pi and just use an ESP32. Get rid of the hosted webserver and use my existing MQTT.
It's better to use switch-case instead of multiple else if :)
Saying that 44.1kHz 16 bit sounds pretty ok is quite an understatement for being literally better than humans can perceive.
Haha true
Why was the output sound so disrupted? I was expecting that i2s is pretty much lossless. I am thinking about creating wireless 5.1 sound system (it will use only power cords) using my old 5.1 (with too much cables) and using esp32 and esp-now. Problem is that I need to convert 5.1 audio to i2s and then distribute that to each speaker. Any ideas? It doesn't need to be esp32, just everything which would make it wireless 5.1 and is already documented.
Your content is really great!. Please make a video on "Commercial grade Low Power IC's (Bluetooth and wifi) for low power IoT". Please compare them with their specs like current drawn while transmission and reception. I would love to get insights about the topic, from you. Thank you!
Haven't played with ESP32 boards yet. I've dome alot with the arduino and the dfplayer mini. Sound quality isn't as good, but works for the small projects.
I've got a couple esp32's kicking around but haven't used them yet... although that's gonna be changing soon since the 8266 has a bunch of unusable ports on it
Thanks for sharing!
Nice going through all that just to end up with underruns
East or West Great Scott is the best ❤️
why don't you repair some Variable Frequency Drives and show us some basics, would love to see something like that from your channel
The other day i,'ve tried to attach the i2s dac to my router. Finally the Linux community have fixed the dma driver and now my router can play music :D
Awesome :-)
@@greatscottlab Yes, my router is based around the mt7628an chip. It have pcie x1, lots of gpios, i2s or pcm. It also have 2r2t wifi 2.4Ghz.
If you run Openwrt you can do a lot of cool things!
500Mhz 2G Sample scope with touch screen how nice :-9. What do you do for a living?
I would love a device that would let you use as an interface for salvaged amplifiers from from broken TVs as they use I2S.
Bro!
You're pretty awesome!
I guess that I have said it before but doesn't matter.
I feel like a failure of an engineer when I watch his videos 🥺
4:02 mediainfo 👍
Could you use the two 8 bit DACs and combine them with the R-2R resistor arrangement to make a higher resolution output? A downside would be only 1 channel, no stereo
You would need r-256r, but yes.
Is there a way to do this as an audio mixer? Like, playing multiple audio tracks and adjusting the volume for each track independently?
Hey GreatScott, can you please make a video on building an audio amplifier using STA380BW
I thought that using an external DAC would improve the quality but I see/hear is not the case...
Good afternoon Mr. Scott, you went all the way to be able to playback a 16bit, 44.1kHz, stereo sound and the quality is appalling like that? That is surely no better than a mono 8bit output. In fact it is reminiscent of audio output of the Commodore 64 when the audio chip was turned on and off quickly enough to transform pulses in actual speech. I have worked with I2S equipped boards that have perfectly good audio quality, that one is unacceptable - for the reasons explained in the several first minutes of the video. I have reasons to think if you used a lower sampling rate and resolution you'd get a better result.
Hello! Can you take a look at the Sonoff WiFi Smart Switch for a DIY or BUY episode?
Where can I get the code for the Arduino at timestamp 8:54?
check your microphone, there's noise all over the video. You are amazing!
What noise?
Its background music I think, you can hear something more than his voice sometimes
So now you can hear yourself saying STAY CREATIVE!!! All the time
とても参考になりました！ありがとう😊
Please make a video on - voltage (-5, -12) . How it works and why it used.
Also have you written the Device Driver for PCM5122 ? How are you providing the I2S clocks ? What is your CLK source ? Can you support 32-bit PCM instead of 16-bit PCM..?
I am using "Node MCU 8266" for my project !!!