Criteria, not a ranking
How to judge one of these apps without trusting anybody
Firstsong is not on the App Store yet, so this page cannot crown itself and will not pretend to rank the apps it competes with. What it can do is hand over the checks we use, so you can run them on any app in this category, and on ours the day it opens.
Our engine, running live on three public recordings. One of them is nothing at all, and it says so.
Firstsong is not a medical device, it is not a doppler and it does not use ultrasound. It cannot tell you that your baby is well and it cannot rule anything out. If you notice fewer movements than usual, call your midwife or maternity unit now. Do not wait to try an app first.
Why there is no winner at the top of this page
Search for the best app in this category and you get two kinds of page. Affiliate lists ranked by commission, and pages published by one of the apps in the list, which somehow finishes first. This is the second kind of page, written by one of the apps, so the ranking is the part we removed.
There is nothing to install or buy from here, which is the only honest reason a page like this can exist. The checks below are written to be used against us as well.
The eight checks, in the order that saves you the most time
| The check | How to run it | What a weak answer looks like |
|---|---|---|
| The arm test | Put the phone on your forearm, a cushion or the sofa and start it | A rate in the fetal range on a limb that has no fetus in it |
| Does it ever say no | Run it three times on a spot where you know there is nothing | A number every single time. Something that cannot fail is not measuring |
| How fast the number appears | Watch the clock from the moment you touch skin | A rate in under five seconds, before any rhythm could be confirmed |
| Can you hear it raw | Listen on wired headphones with any enhancement switched off | A crisp loop that sounds identical in every session |
| Is the price written down | Look for a number before you install, not after | Price visible only inside the app, after you hand over an email |
| Does it claim doppler or ultrasound | Search the store listing for both words | A microphone app calling itself a doppler. That is different physics and the developer knows it |
| Does it ask what week you are in | Watch the onboarding | No question at all, and the same promise in the first trimester as in the last |
| What it does about your own heartbeat | Hold the phone against your chest instead | Silence on the subject, or a number that is exactly double your pulse |
The first check was invented by the users, not by us
Across 1,995 public App Store reviews of eighteen apps in this category, 41 reviews describe a woman testing the app on something that cannot possibly contain a fetal heart. A husband, a child, a dog, a bed, a sofa, a floor, an arm, a thigh, a hoodie. 31 of those 41 are one star. Nobody taught them that test. They worked it out because the app was too willing to agree with them.
"My bed has a heartbeat."
"I sat my phone on the floor and recorded for 10 seconds. I guess my floor has a heartbeat suddenly."
And the five star version, which is the one worth copying: "I put it on my leg to make sure it wasn't detecting a heartbeat there, and my arm. No sounds. Then I listened to my own heart, clear as day."
The largest app in this category answers it in a support page of its own, under a heading that reads Even my sofa has a heartbeat!, beginning "We get this a lot! :)" and containing a paragraph headed "SOFA BUG UPDATE!". A second heading asks "Even my boyfriend's tummy or leg has a heartbeat?". Those are their own words on their own site, and they tell you the check works.
The strangest number in this category
Of those 1,995 reviews, 574 state how far along the woman was. Sorted by that, the ratings run backwards.
| Gestational age in weeks | Reviews | Five stars | One star | Average |
|---|---|---|---|---|
| 4 to 12 | 150 | 63 per cent | 25 per cent | 3.79 |
| 13 to 19 | 163 | 41 per cent | 44 per cent | 2.98 |
| 20 to 27 | 115 | 13 per cent | 70 per cent | 1.90 |
| 28 to 42 | 146 | 12 per cent | 79 per cent | 1.62 |
Satisfaction runs precisely opposite to the physics. The earlier in pregnancy, when there is least for a microphone to pick up, the happier the reviews. The stretch where a phone has any chance at all carries the worst ratings in the set. Whatever those early five star reviews are hearing, it is not a baby, and the app agreed with them anyway.
That is why check seven matters. An app that never asks what week you are in has decided in advance to say yes to everybody. The week by week version is here.
Run check two on us, right now
The third recording in our tool is real audio with no rhythm in it. That is the input that produces the arm and sofa reviews, 41 of them in a corpus of 1,995. Watch this one decline, in your own browser, before you believe a word on this page.
Nothing is uploaded. No account, no cookie, no request to any other domain.
What a checkable claim looks like
Any app can say it works. The difference is whether the claim arrives with a number, a public dataset and a way to reproduce it.
The limit belongs in the same paragraph as the boast. Those recordings came from research sensors on the belly, not from an iPhone microphone against real skin. That measurement does not exist yet, for us or for anybody else here, and until it does nobody deserves trust on it, including us.
Price is a check too, and here is where we sit
Of the same 1,995 reviews, 373 mention price. 294 of those are one star. People in this category do not complain that apps are expensive, they complain that they paid for something that did not work.
| Annual price across 28 apps with a readable subscription | Amount |
|---|---|
| Lowest in the market | 8.99 USD |
| Median | 19.99 USD |
| Most common single price, across six different developers | 39.99 USD |
| Highest anybody charges | 69.99 USD |
| Firstsong, planned | 79.99 USD |
So we sit above the entire market, and we would rather write that down than let you find it. An app that convinces you it works and an app that refuses when nothing holds cost the same to build, and only one of them is worth paying for. There is no trial either, because a trial that renews into a yearly charge is how this category earns its worst reviews. The first real listen happens before you pay.
Four things you should hold against us
- Firstsong is not in the store. You cannot run these checks on it today, which is a reason to be skeptical rather than patient.
- The number that matters is missing. Our results come from research sensors, not from phones on real bellies.
- There are no clinicians on this team. Nothing here is reviewed by a doctor and no page on this site claims otherwise.
- There is no launch date, because the date depends on a measurement that is not finished.
One email, the day it opens
Firstsong is not on the App Store yet. Leave an address and you get exactly one message when it is, with the price on the first line. No newsletter, no drip, no partner offers. You can reply to that email and a person reads it.
Stored on our own server, never sold, never shared. Your IP is hashed, not kept. Nothing about your pregnancy is asked for or stored.
Common questions
So which app is the best one?
We cannot answer that honestly, because we are one of them and we are not even released. What we can tell you is that in this category the checks above separate the field faster than any review score, and that the first one costs you thirty seconds.
Are the free ones worse?
Free and paid fail the arm test at about the same rate. Price is not the signal. Whether the app is capable of saying no is the signal, and that is free to test.
Do star ratings help?
Only if you read the one star ones and count how often two sentences repeat: it did not work, and I think that was my own heartbeat. In the corpus we read, waste of money appears 173 times.
What about apps that say they use ultrasound?
A phone has no ultrasound transducer. An app on a normal iPhone cannot emit ultrasound into your body, so a listing that claims it is either using the word loosely or is describing something the hardware cannot do. Both are reasons to close the page.
Would a home doppler be better?
It is a different trade, not a straight upgrade, and midwives have specific objections to home use that are worth reading before you buy one. We wrote them up here.
When can I run these checks on Firstsong?
The day it opens, and the email list is the only place that day gets announced. If the phone measurement comes back badly, that same email will say so instead.
Keep reading
- Do these apps work. the same corpus, read as one question
- The arm test. check one, in full, with the reviews that invented it
- Is it my heartbeat or the baby. check eight, and why the arithmetic makes it so convincing
- Are these apps safe. the risk is not the sound, it is what you do after it
Last checked 2026-08-23