Hacker News new | past | comments | ask | show | jobs | submit login
Show HN: Sawtooth – Online audio workspace (sawtooth.io)
109 points by myzie on March 29, 2017 | hide | past | favorite | 42 comments



I find myself asking questions that could easily be answered with a demo or some sort of try before signup.

The idea of everything I click on going to the signup page really is a huge turn off and personally I consider it to be a poor user first experience.

You should add some sort of something that allows me to see what you are about before making me sign up. I don't even know what I get if I sign up.

Otherwise, I am sure you have done a great job.


Same UX here.

"Oh a play button, audio sample maybe?" click "Sign up!"

No thanks. Just not a good user experience - I want to know what I'm signing up for. Some screenshots, a video, or at least some audio samples of what it can produce. Something other than a text feature list.


Thanks, you are right. I should add more content up front so that everyone can get a better idea how it works without having to sign up.


Even better - the home page should be the app itself. Only prompt people to register for an account when they've created something they might want to save.


This is what I was expecting.


Yeah, comes across as a landing page for a pure lean-startup approach. Capture emails to validate this is something people are curious about.


As a frequent user of SoundCloud, Gobbler and plain old DropBox to share audio files among fans and colleagues in the industry, I am wondering what the advantage of Sawtooth could be?

I see that you have filters etc., but given that I would rather adjust EQ and effects on my tracks on my own DAW with virtually no latency, I don't know that I would actually do that on a web platform with all the vagaries of lag, dropped signals etc.?

Plus the fact that most audio people have their favourite 'go to' plugins for reverb, delay, chorus etc. as VST/AU plugins that they pull into their DAW - it seems that Sawtooth straddles that line between being a quick 'grab an audio recording snippet for sharing' and a full fledged web based DAW.

I am assuming a 'use case' for this would be to capture a song or riff idea while I was sitting in a hotel room between travelling etc., but to be honest, I have a lot of iPhone apps for doing that and posting directly to the sites I mentioned above. To make Sawtooth compelling, it would probably have to supply some rudimentary DAW like capabilities, such as perhaps a metronome, some sort of ability to do basic MIDI patterns with uploaded samples, and perhaps some rudimentary multi track ability - even 3 or 4 tracks would be great for doing basic song ideas to send to my band members.


Gobbler looks great. Have you used the collaboration features and have they worked well for you?

One of the goals is to streamline the experience of navigating and listening to sets of files. If it's built correctly, then using Sawtooth should be a smoother experience than managing where your files live in your Dropbox folders, having it sync them, finding them on another computer, then listening with a separate media player or the player in your OS file navigator (thinking of Finder on MacOS).

This is not meant to replace any of your desktop DAWs. If anything, it could interoperate with them in certain cases, if people are interested in that. Maybe an API would be handy for others developing websites or apps that work with audio.

I'd like Sawtooth to keep simple audio work really simple. I suspect it's easy to overwhelm newcomers to the audio world with complicated UIs (which are necessary for advanced work).

Gobbler for example seems very geared towards musicians and music creation, which is great for many. But there are also lots of people working with audio for other reasons... field recordings, voice recordings (podcasts etc), signal analysis, etc. Maybe Sawtooth becomes more optimized for one of those other cases.


As an amateur producer, I'm wondering: what's the utility of something like this?

If I create a new set, I have the option to use the 'Synth'.

Let's be very honest here: if I want to make real music, I'm going to turn to a serious DAW + synth plugin. I personally use both Massive and Serum with Ableton. Anything you cook up in a webapp is going to fall seriously, seriously short of what Serum can do.

Not to mention that a web tool just doesn't fit into the workflow. The synth is the heart of digital music production. If I'm making music in Ableton, I want my synth to be inside Ableton.

This might interest absolute amateurs, but amateurs won't pay for this, and by the time they are advanced enough to want to pay, they would have discovered Ableton/Logic/FLStudio.

I don't mean any offense, but I find that online music tools like this are generally very poorly thought. They are a solution searching for a problem instead of the other way around.

A lot of people need simple tweaks to their photos or graphics. This is why online photo editing tools work even though they fall far short of Photoshop in capabilities.

But audio/music? This isn't something your average Joe needs for his Instagram profile or his Facebook business page. If someone is serious about music/audio editing, he will eventually want to use a professional tool.

Not to mention that Ableton Lite is quite decent for someone new to music production


Sawtooth creator here...

Thanks for the feedback and I agree with a lot of what you said. You're right that this can't compete with pro audio software but it's also not my goal to compete with those tools.

I'm putting Sawtooth out there to evaluate if there is demand for this type of web app, and what groups may be the most interested. Whether anyone will pay for it... great question! I'm not worried about that just yet.

Web apps have a lot of limitations compared to native desktop apps when it comes to serious audio work. At the same time, I wonder how many people could use a reliable web app for super quick edits, to listen to some of their audio while on the go (not just at an audio workstation), or share tracks privately with bandmates or coworkers. We'll see.


Thanks for your project.

The sharing aspect is the interesting bit. Having a "github for sound samples", with the ability track forks and have commit history could be wonderful.

Implementing a professional DAW or synth takes a staggering amount of work. The toy synth might confuse users about the purpose of sawtooth. (Personally, I would remove it)


See: Splice.com. It works like GitHub and is already very popular with producers.


I'm glad you are taking all the feedback in stride but I think this is a great idea. For all the people saying signing up is a big commitment, buying or ripping off (as thats what everyone I know did in college who couldn't afford it) Ableton Live or for me using fruity loops or a linux knockoff is a big commitment either financially or learning the ins and outs.

At the very least, this takes away some of the

a. intimidation of merely experimenting with music to someone curious about it, and could be a leeway to people investing in desktop based software or a group of users could grow with you as you go along.

b. financial commitment or technical search/compare contrast, and download/ripoff of existing software and finding compliance with operating systems.

Most of your feedback here is probably going to be from experienced music makers but as an Electrical Engineer who loves music, and playing with music is only something I have to do in my spare time since I work two full time jobs, this is really cool and allows me to play with stuff easily.

It also enables me to login and work on this at my work computer, so even if I make a sample, I can send it to myself elsewhere.

I think this is a great idea. There is some cool music software and I have audactiy, ardour, ableton live and fruity loops but the webapp idea is actually a great one and there are some huge advantages here. You are

1. Lowering the barrier for entry level music makers and making it a little less intimidating to be honest

2. Allowing increased access so people can play and edit in their spare time or on a lunch break at work, or just pull it up to play around with friends at a coffee shop on their laptops.

Going through the process of finding downloading/paying for a piece of music making software is alot of work and might otherwise deter two people from sitting down and just mixing up something on the fly quickly

3. I know alot of highshoolers who would use this in school that can't bring their laptops to school.

4. On a globalscale, feedback here is really underestimating the lack of ability many people in many countries have to liscenses to good music making software we have in Westernized society. Now anyone who has internet or the future phone apps can have access to this without having to torrent software or be a linux guru to know about the free distros out there. This is a huge advantage.

5. As shown in another comment, someone who does speech processing has already found this to be advantageous, there are some clear uses outside of experienced DJs and music makers for other industries. Watch your demographic closely because there could be some great uses for this otherwise that come through in later funding rounds if you need support to scale.

6. The basic features despite immediate feedback from people that it seems pretty basic, is not a bad thing when you are starting out in my opinion. You can grow and respond better with your user group than having a panel of overhwleming options to anyone other than already experienced music makers. Their feedback is one of the most important, but I think your initial userbase is going to grow with you and that's not a bad thing.

Like everyone else said, I would put a cool little mini walkthrough tutorial of how to make a cool 20 second sample with your software and have it on the front page so people could actually click through see some of the features and make a little sample or something. If they want to save the sample to edit, share or collaborate, then they can be prompted to make a login.

When you roll out to the paid app, provide a discount if people to invite, share on twitter, fb etc and provide a link for them to do so. It sounds cheesy but don't be ashamed to have your own users advertise.

Don't be discouraged by the feedback of experienced users. This is a really good idea.

scale for phone apps and make a twitter account.


I really appreciate hearing your perspective on all this! Lots of good points that I will consider.

The convenience factor and lowering the barrier of entry is definitely what I thought would have some appeal. Use in various education settings included.

Edit: one thing that I realize now is not clear at all -- anyone with a Google account can use it to login (no password) if you visit the "sign in" page instead of "sign up". I need to make this more clear.

It's definitely apparent now that having a way to see how it works without having to login would be a great improvement. I'd considered it before but haven't gotten to it yet since the implementation will be some work. I'll reevaluate.

And thanks for the other tips.


> If someone is serious about music/audio editing, he will eventually want to use a professional tool.

Perhaps, but there are a good number of people that need to do very basic audio editing frequently and don't really need something with more features than this.

An example: I'm part of an online community that focuses on video and film preservation and a good number of its members spend time capturing commentaries from Laserdiscs that haven't made it to DVD or BD. These captures need to be synchronized to the track on the target release in order to be enjoyed. Only very basic tools are needed for this work.


I think this is more of a editor like Audacity that just happens to have synths. I don't see any sequencing features.


I think the marketing page could do with some demos. There are online DAWs already and most of them are pretty sad. If you showcase some real audio work accomplished with Sawtooth, it'll be more worth signing up to try.

I have DAW software already, I use Ardour and Pure Data on my computer, so I want to know that I can do something compelling with your tool before bothering to set up an account.


I appreciate the feedback. A couple aspects of Sawtooth that might be compelling for you, as a supplement to what you're already using -

1. Share sets of recordings privately with any collaborators. All protected by logins. This feature is basic right now (it's read-only if you are not the owner) but could be expanded.

2. The convenience factor of accessing your most used audio clips or recordings from anywhere, on most any device. Maybe your final Ardour mixes you would upload to have easier access when you're not at your workstation, for example.

I know this isn't for everyone, and it's certainly not intended as a replacement for desktop pro audio software. Rather, it may be a complimentary tool for some for simple editing tasks, quickly streaming your recordings while on the go, or collaborating with others on shared sound files.


Seems cool, good job. I look forward to seeing more of your platform.

DAW plugins are usually pretty hard to write and maintain, but maybe it'd be cool to have a "send selected tracks to sawtooth and share" type flow.


I've very long wanted something like this! Basically like a CyberChef ( https://gchq.github.io/CyberChef/ ) but for audio. I work in digital speech processing and often work with audio codecs and it would be very helpful to have an online tool which lets you apply multiple filters to some audio.

What I'm missing after trying out Sawtooth is the interactivity of CyberChef. Basically trying out a few filters, one after the other, showing the intermediate results for each, and some frequency analysis on it, and hearing the results. Can audio filters be applied in the browser for instant-feedback to hear the audio 'live' (like CyberChef auto-bake) like it can be with text?

And some conversions would be really great as well, like being able to apply various encodings: Treating the samples as a-law encoded and applying an a-law decoding pass on it.


I was wondering if anyone would be interested in having some signal analysis available. It would be a logical step from what's there now.

At the moment all filtering is done in the backend. Hearing instant-feedback though seems to be a common request so I'll have to give it some thought.

Sawtooth does keep all versions around and available (to support undo) so improving the ability to quickly play the different versions of the same file would be the easiest addition.

Thanks for giving it a try.


Did you ever have a desktop app? In the early 2000's I used a program I thought was called Sawtooth, or maybe it was just SAW, 'saw' was in the name, but I can't for the life of me find it anywhere. You could draw a wave form with the mouse, name it, piano roll it, sequence it into a song and export as a wave and that was about it. I know that sounds like ever DAW on the planet but it didn't have any advanced features and I think drawing the wave form was unique to it at the time. If anyone knows what I'm talking about please let me know.


2 days of really hard thinking I finally remembered. "SawCutter". Seems to be abandoned though. cuttermusic.com is something else now and the download.com link is, of course, completely shady. Hope I can find it on an old backup CD somewhere.

Edit: Looks like this might be the author who seems to have a pretty impressive resume: http://www.larryzitnick.org/


Not me! Doesn't ring a bell.


I do a bit of online video production (for social media marketing and for online courses) and I have to open up audacity every once in a while. I don't like the audacity user experience. Here's my feature request list

1) Make it easy to switch audio formats ([mp3|wav|au|etc]->[mp3|wav|au|etc])

2) One button to make it louder, one button to make it quieter

3) A good cut and paste interface with the ability to zoom in and out and see the spectrogram so one may be sure that one starts cutting at the audio part (if there is white noise)

4) Abilities to selectively remove deep or high voices and remove background noise


Thanks for the list!


You're welcome. I know how challenging it can be to launch a new product and all you want is information from the market regarding where to go.


Interesting. In "Works With Multiple Formats" section, it mentions 'AU' - do you mean AIFF (AU is Apple's Audio Unit plugin format)? 'Chorus' is misspelled under "Filters".

What is bitrate/quality of transcoding, and how would you rate your DSP algorithms compared to those in pro DAWs/editors? Like, is that FreeVerb or something fancier, and what about pitchshift/timestretch quality and zero-delay filters? Audio folks are pretty picky about quality these days.


In supported browsers I'm using the Opus codec which in general is quite good quality compared to MP3. It falls back to MP3 in browsers that don't support Opus... Safari and IE I think.

I believe it's using the default encoding settings for Opus and MP3 at the moment (using opusenc and... maybe lame for MP3 I forget). Certainly I'm looking to have great streaming quality so I should confirm that those defaults are reasonable.

I'm using various open source and custom tools for the processing. YMMV. In general they should be solid, but not as fancy as many of the latest VST plugins. This could all evolve depending on feedback and what people are interested in. One thing I considered as an addition is the ability to define custom filters on the webpage... either interpreted or compiled in the backend to edit your files. I think that would be a neat way to experiment with filters, but would have some limitations as well.


It's an old PCM format from Sun

https://en.wikipedia.org/wiki/Au_file_format


I like this idea - but why do you let people edit the .wav part of the extension? That was super counterintuitive to me could I have done .mp3 for example?

I'll leave any other feedback I have here unless you have an email I can send it to - I've been looking for a good web version of Hum (the mobile app) and this looks like it could be it + more!


Hey there, thanks for trying it out. Was this after uploading a wav then clicking to edit its tags or file name? I'll see if I can improve that aspect. In general, Sawtooth transcodes uploads behind the scenes to create mp3 and opus encoded versions so that they can be streamed to your browser (not all browsers support playing wav directly). These versions are all stored next to each other. Editing the name string in the UI shouldn't be able to change file extensions in the backend.

I'd be happy to field any other questions here or equally you can reach me at curtis at sawtooth.io


Seems like a great start. I too would like to see ways to combine waveforms. And a bigger ask of being able to draw filters in real time while playing back to hear changes instead having to process them to find out what will happen as well as loop the same while editing. I realizing I'm asking for more DAW like features! :P


Seems like a really cool project with potential. Would like to see the ability to preview on filters and synths, that seems like a big downside compared to the desktop DAWs.

Also not sure if there's a place to report bugs and provide feedback in the future?


Thanks, please send any feedback to curtis at sawtooth.io


I was playing with it for a bit and came across a few issues, but can't seem to find a support or contact link. Where would you like such information sent?


Please shoot me an email with your findings: curtis at sawtooth.io

Thank you!


It would be worthwhile to mention if multichannel audio is supported or if the application is limited to working with mono and stereo inputs.


Good point. Multi-channel is supported. I'll make a note to add it to the feature list on the page.


FYI - you can use Google login if you visit the Sign In page (instead of Sign Up). Need to make this more obvious.


Click - sign up. ... No.


Can It be used for direct streeming?




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: