The Voice AI landscape is rapidly evolving, with numerous platforms offering developers pipelines for integrating voice, large language models (LLMs), speech recognition, and transcription capabilities. The convenience of ready-made pipelines is undeniable, however a concerning trend is emerging – a race to the bottom driven solely by pricing.
I recently consulted a team that seamlessly migrated their infrastructure from one voice AI platform to another within a mere 1-2 days, motivated by better pricing and experience. A week later they stated they would build their own starting with the lower hanging fruit again to lower prices (volume pricing wasn't enough). This eye-opening experience underscores the commoditization of these platforms, where the overall experience is largely similar, and pricing becomes the primary differentiator.
This raises the question: Is this a race to the bottom? If so, should voice AI platforms start by undercutting the market with competitive pricing, and then focus on expanding their offerings and identifying innovative monetization strategies?
Another problem I see with most voice ai pipeline platforms is the "one-size fits all approach," and lack of customizations. As a result anyone in the Voice AI space may leverage the pipelines to prove value propositions but leave long term. For example - majority of tools only allow one large prompt. This creates an issue because the one prompt is responsible for many pieces which creates bugs and overemphasizes prompt engineering and large prompts which is horrible. Rather mini prompts can be issued and dynamically changed building a more customizable solution built from scratch but no one is offering this right now.
However, initially winning over users through competitive pricing, these platforms could amassing a substantial user base and gain valuable insights into real-world use cases, enabling them to develop more tailored functionalities and better products. Also, voice AI platforms could position themselves as prime acquisition targets for enterprises seeking to bolster their developer ecosystems. Rather than engaging in short-sighted price gouging, a more sustainable approach would be to leverage the volume of users to explore creative monetization models and continuously refine their products based on user feedback and emerging trends.
These are just high-level thoughts.
]]>Reddit is the one place for the highest authentic feedback on anything. Its unbiased, raw, filtered and people just don't give about how you feel when it comes to their thoughts and input. You can literally research user feedback on your product, your competitors product and identify opportunities in the market place. One major hack when researching and hyper focusing on where to find exactly what you want using Reddit is using Google Search and typing in
site:reddit.com "what ever your looking for"
This will identify all reddit posts / forums with the keywords you're looking for or anything like it.
A couple other things I do:
As for other places with similar feels. Facebook is a closed platform so its not easy to find them. Twitter is now closed as well and generally comments in social platforms where identity is exposed aren't as good but worth a shot. Other platforms may be better depending on the age groups and demographics. Just figure how how to hack the search so you can research effectively on the other platforms. Click Here to try out the search
God Speed;
]]>Nearly all voice applications translate speech to text. However, while voice applications need to do multiple things analyzing text we still return to the audio itself for deeper processing.
For example we can analyze text to get intent analysis but we need to analyze the wave patterns of the audio file itself to get "real" sentiment analysis. But I wonder if that is the wrong approach.
I am learning Spanish and right now I translate Spanish words I hear into English for further processing. This is a multi-step process that is time consuming and delays my response. My friend Yoni at Mister C's Barbershop dealt with this issue when he had to pick up English coming from Spanish. He said he didn't really learn English until he challenged himself to stop translating in his head and instead associated the words with the objects itself. At that point he could skip translating and respond extremely fast because he essentially learned the language - we do this with our native tongue.
What if we do the same approach using audio in voice ai applications. We could remove translating speech to text - understanding what is being said by analyzing the audio itself and comparing it to pass records using a vector database and pass records of audio, translations, intent, speech to text, etc. If we don't have a similar record translate speech to text (aka inquire).
If this were to work data is once again at the top of the list and will significantly reduce cost and time required for voice ai applications to understand and respond.
What is also interesting is it follows the human way of learning. We need to be exposed to things, directed and corrected for a certain amount of time to understand languages.
My final quick thoughts:
We are in the early days of voice ai protect your data. Companies with the data will significantly outpace those with little. Speech to Text should be an beginning step. Keep the raw audio. Cut the raw audio into clips and pair them with the translation, sentiment, intent, etc. Storing the raw audio into a vector database for it be queried based on similarity.
The new trend is Voice AI and I believe "voice first" will be how we build applications with a fall back to traditional methods. This is great however, I think the approach of building by relying on large single prompts, letting AI run free based on set instructions within prompts, is wrong. Not to mention current costs range from 12-25 cents per minute which means while it scales well, most AI Voice methods cost more than having an overseas representative.
Running Voice AI apps specifically for phone support is expensive, especially if you lack scale to reduce costs through volume, didn't optimize through engineering, or build from scratch without a thoughtful way to divide problems. Right now most building in this direction rely on AI to handle everything and provide the right outputs. The results have been subpar and over-engineered in the wrong direction.
One specific use case is AI versus IVR. Do we really need an entire AI system just to replace the IVR, or can we merge the two?
I've worked with IVRs for Autodial.com and noticed most inbound call apps don't need complex voice AI. While IVRs are "bland", they've cost-effectively solved most customer service issues for decades compared to expensive AI. The problem wasn't the IVR itself - it was wait times and reaching a real representative.
When building, we want to effectively solve customer issues, not just add sweet frosting to a bad cake.
For years IVRs have been used for:
But IVRs have faced issues:
An intermediate step before complex AI could be Smart IVRs combining the best of both to resolve limitations and keep costs low while allowing more complex capabilities if needed.
A Smart IVR should:
I'd argue Smart IVRs are the way to start effective AI voice systems for inbound calls. With the ability to jump around, micro models can be easily implemented and optimized over time, reducing costs and improving performance. Voice applications are here to stay and costs will fall, but we should consider the best way to blend IVR and AI to solve problems.
]]>Knowledgeable or Work-Ethic?
Which do you choose between the two?
One can get the job done.
The other will get the job done.
Of course we want to optimize for both.
However if you ever doubt your abilities
Remember you can work your way up.
The person with the work ethic will outperform the knowledgeable one.
and over time the person with the work ethic will rise to the level of knowledge required.
Apple's 3D Touch is a valuable tool for applications and developers. However, a notable omission is the direct access to an app's specific settings via 3D Touch. Currently, options like "Remove App," "Share App," and "Edit Home Screen" are available, but integrating a shortcut to the settings would optimize the user experience.
Currently navigating to an app's settings requires users to open the Settings app, scroll to locate the app (a task given the average user has around 80 apps installed, potentially even more in the US), and then access its settings. This process involves at least five steps compared to the one or two steps a direct shortcut would offer.
Another beneficial addition would be the ability to use 3D Touch on an app's notification to quickly access its settings, particularly to manage push notifications. Providing easier access to app settings could also encourage developers and businesses to adopt more user-friendly practices, knowing that users have the ability to quickly adjust permissions and settings. This feature could influence the ecosystem to be considerate of spam notifications.
Context:
]]>I recently implemented dark mode for the second time ever on Autodial, with my first here on this site (toggle sun/moon button in top right of the header). The first attempt was complex because I didn't fully understand the CSS, HTML and JavaScript components. This time was much easier by following three key steps:
Enabling dark mode enhances user experience, even if minor. I'd argue enhancing user experience will increase profits and companies that don't offer dark-mode likely don't prioritize UX. A drag-and-drop solution or GitHub library could help automate dark mode for many sites. Overall, implementing dark mode is straightforward today and should be considered standard practice
]]>In the past, I was very dismissive of Progressive Web Apps (PWAs). I felt they were slow, buggy, and provided poor user experiences compared to native apps. As an iOS developer, I'll admit I was biased against them. However, recently my views have changed. Web technology and browser support for PWAs has improved dramatically.
PWAs now offer several advantages over native apps.
Apple doesn't make it as easy to install PWAs, hiding the install button and restricting JavaScript installation. However Android is extremely easy and you can prompt the user to install. Apple's reluctance is likely because PWAs compete with their lucrative App Store business model ($85 billion in revenue in 2022). But the technology has advanced enough that PWAs are now a viable alternative for many use cases.
Overall, PWAs have improved and startups should consider them before building a native app. For basic functions, most apps don't need to be full native apps anymore. PWAs provide a faster, cheaper way to get to market while still providing a quality experience.
Additionally, having a Progressive Web App should be strongly considered if you are providing a web browser experience. The incremental effort to make an existing site installable and app-like is typically quite low. And it allows users to access your product in more seamless app-style way. The most important push for this is there is little to no additional costs to making your website (mobile responsive) a PWA unless it was built wrong and there are plenty of plugins to make it easy.
However, there are still some cases where a native app may be preferable:
Majority of the apps we use also have a PWA option:
Startups should give serious consideration to beginning with a PWA as it allows faster iteration at lower cost and similar high quality experiences. For people and companies starting out - if you want to provide a website experience to your users or already have a website experience you just need a PWA. Please spare yourself the time, money and headache. While PWAs now offer a viable alternative in many cases, companies should evaluate if their specific product needs warrant building a full native app instead.
Below are a list reference links for PWA's
I want to start posting ideas and concepts big or small. Ideas don't mean anything, execution does however I want to provide a space to push out ideas big or small. Maybe someone can pick it up and run with it, maybe I will try them out myself. This is a live list.
Develop a comprehensive list of companies that operate product directories, offering a centralized platform for product submissions. Organize this list into intuitive categories and make it available at no cost for individual use. Additionally, provide a convenient service where, for a fee ranging between $20 and $50, companies can opt to have their product submissions handled by our team through a single, streamlined form. This fee is competitively priced, with $20 being an accessible entry point. Companies also have the option to self-submit to these directories without any charge. This list and service will be hosted on a user-friendly website, optimized for easy sharing across social media and various digital channels. The promotion strategy includes directly reaching out to the founders of prominent startup directories such as Y Combinator, 500 Startups, and Tech Stars, using targeted email marketing.
What is interesting is it could become so big that eventually it becomes another product site and possibly "the" product site through utility of working with other product sites.
Then provide insight for how people could market their site based on the type of product it is which could create other revenue opportunities like marketing consulting services.
Verdict: Not Tested
Below are a list of quotes, thoughts and life lessons I live by. This is a living document since I'm constantly learning and remembering lessons.
Doing a good thing at the wrong time is still doing the wrong thing.
It's not about getting things done, it's about getting the right things done.
Doing right thing doesn't come with your emotions.
If you fail to plan you plan to fail.
Don't ask permission, ask for forgiveness.
Respect execution, not ideas or intention.
Money only uncovers the inevitable with delay or speed.
Actions speak louder than words.
How you do one thing is how you do everything.
I was going to = but you didn't.
Be humbly bold and confidently arrogant.
No one cares unless you're pretty, rich, or dead.
Almost never counts.
The good that happens to you is your fault and the bad that happens to you is your fault too.
Fighting is for losers.
Everything seems to works this way.Sports work this way - Dan Gable, Kobe BryantEmpires - Roman and CeasarPhilosophy - Marcus AureliusBusiness - J.P. Morgan, Steve Jobs, Warren BuffetInstitutions - TeslaScience - Newton's law of gravitation
As long as you achieve something amazing that pushes things forward and it continues to be relevant, you can be remembered. The thought of living on is counter intuitive. I don't think they aimed to live on forever; rather, I think they aimed to do great work and go deeper and deeper into their work. Maybe the better process is to acknowledge, comprehend and get to work.
The question isn't do we want to live on. Maybe it is do we want to do meaningful deep work and not stop; and if so, where can we grow to care about enough we want to continue doing deep work and high amounts of contribution in. Where can we contribute that has an endless game even after we pass.
Will your name live on?
Even if it means slimming things down from your ideal version 10 to version 1.0 or 0.01. Finishing things is a real skill that we want our name to associate with. It protects our limited time, energy, and resources. Half done / almost isn't it. Would you rather work with someone who gets it halfway done or someone who finishes what they set out to do every time? It's better to work with a hard-ass that finishes things than the fun person who doesn't.
Protect your time work with the finisher.
Even in the situation where you don't care and you just want to have fun...do both but most importantly finish.
Protect yourself, be a finisher.
I spoke with my performance coach,
and we agreed, I must choose (wisely)
where I place my focus, and who I assist.
I truly love to offer help,
but time and energy have limits...
as do all of ours.
Often, I invest so much,
only for it to be squandered,
or to foster dependence,
where I'm always required (not ideal).
My task is not to aid everyone,
but to support those who matter to society,
those who construct grand endeavors.
If it doesn't aim to disrupt,
dominate markets,
and alter our culture,
it can't be my concern.
I prefer those who build Turo,
rather than those who leverage Turo.
I prefer those who build Airbnb,
rather than those who leverage Airbnb.
But progress occurs in stages,
small steps, not often massive leaps.
So, how do I discern which idea and who counts?
If money is a report card,
its thousands & millions vs billions & trillions.
But the question hinges on the goal.
Distinction lies in purpose.
If it's wealth, that's one aspect...
but forging a new (or improved) method,
that's intriguing; it changes the current flow.
Imagine ants in a long line on a tree,
some ants convey a new route,
or hint at another option...
This revelation would interrupt the ant line.
Informed ants take the novel path,
while others stick to the original,
Until it ceases to be logical.
Relating the ant metaphor to earning more,
is akin to an ant lifting a large load,
yet following the same route.
It benefits the ant, maybe a few
but the flow stays the same.
One approach serves self,
the other serves the group.
I favor the collective interest.
Perhaps nurturing the group means self-care too.
Maybe wealth will trail that success.
Perhaps that's what investors seek as well.
Someone crafting a superior or novel method,
bold enough to share it with others,
with the ability to change the change flow
and as it succeeds, wealth ensues.