Dictation App Store Ratings: How Many Each App Has

Apple's rating counts split the voice market into a few giants and a long tail. General assistants such as ChatGPT voice mode have 11,431,550 ratings on the United States App Store ¹. The best-known dedicated dictation app, Wispr Flow, has 17,618 ². Most other dictation apps sit in the dozens or hundreds. The dataset holds 540 data points ³, one per app, store and country.

One caution up front: this page is a snapshot from October 19, 2026. A rating count is a running total, so the table shows size, not speed. Growth appears only when the weekly readings are compared over time, and a single day cannot show it.

General assistants have ratings in the millions, dictation apps do not

The top of the table is not dictation software. Amazon Alexa has 5,919,353 ratings in the United States ⁴. Gemini has 2,354,196 ⁵, Grok 1,519,656 ⁶ and Microsoft Copilot 1,360,110 ⁷.

These apps are general products with a voice mode inside. Their counts reflect everything people do with them, not dictation. They show how many people already talk to their phones, but they are a poor guide to which dictation tool is popular.

Siri is the oddity. Apple's own assistant has 1,469 ratings in the United States ⁸. Siri ships with the phone and is rarely opened from the store, so people have little reason to rate it, and its count is not comparable with downloadable apps. In Brazil it has 95 ⁹. Built-in tools undercount their real use.

Dedicated dictation apps are small, and one stands apart

Among apps built for dictation, the gap to the leader is wide. Here are the United States iPhone figures for the segment:

Showing 10 of 18 rows. See every row in the dataset.

Wispr Flow, at 17,618 ratings ², is far ahead. Willow Voice follows with 1,066 ¹⁰, then Superwhisper with 844 ¹¹ and Typeless with 762 ¹². Monologue has 297 ¹³ and Aqua Voice 98 ¹⁴.

The tail is thinner still. Spokenly has 63 ¹⁵. Codictate has 0 ¹⁹, and Amical also has 0 ²⁰. Many new dictation tools launched on the Mac or the desktop first, and their phone listings are recent or lightly promoted.

Meeting and transcription apps have the biggest specialist audiences

Outside the general assistants, the largest counts belong to meeting and transcription products. Otter.ai has 85,529 ratings in the United States ²¹. Plaud, the app for a recording gadget, has 23,677 ²², and Granola has 14,567 ²³. Fireflies.ai has 5,310 ²⁴.

In transcription, Notta has 27,200 ratings in Japan ²⁵, Riverside has 20,179 in the United States ²⁶ and Rev has 4,332 ²⁷.

Voice-note apps sit in between. SpeakApp AI has 9,543 ²⁸ and Voicenotes 7,154 ²⁹. Just Press Record has 1,525 ³⁰.

Voice keyboards carry the weight of old, installed products

Keyboards with voice input have large counts because they have been on the store for years. Microsoft SwiftKey has 121,807 ratings in the United States ³¹, and Gboard has 60,397 ³². Newer entrants, such as Google AI Edge Eloquent with 25 ³³, have barely started.

Age explains much of the pattern. A rating count accumulates, so an app that has asked for reviews since the early years of the App Store will outrank a better product that launched recently.

The same app looks very different from country to country

The United States storefront is usually the largest, but not always. Otter.ai has 85,529 ratings in the United States ²¹, 14,740 in the United Kingdom ³⁴, 6,091 in Japan ³⁵, 546 in Germany ³⁶ and 435 in Brazil ³⁷.

Wispr Flow shows the same shape: 17,618 in the United States ² against 2,628 in the United Kingdom ³⁸, 1,820 in Germany ³⁹ and 174 in Japan ⁴⁰.

Local favorites reverse the order. Notta has 27,200 ratings in Japan ²⁵ but 1,454 in the United States ⁴¹. Accurx Scribe, a clinical tool, has 17,752 in the United Kingdom ⁴² against 330 in the United States ⁴³. Granola is stronger in Brazil, with 1,042 ⁴⁴, than its size elsewhere would suggest.

Professional tools have few ratings, and some have poor ones

Medical and professional dictation sells to organizations, not through the store, so the counts are small. Heidi Health has 2,563 ratings in the United States ⁴⁵. Dragon Anywhere has 726 ⁴⁶, with a low average star score, and PowerMic Mobile has 181 ⁴⁷.

Ratings from these tools tell you little about quality in a clinic. A hospital that licenses a product rarely asks its staff to rate it in the store.

Where studies and sources disagree, and what the data cannot show

There are no studies to compare here: every row is Apple's own reading on one day. The disagreements are between ways of reading the same number.

• Counts are not users. Ratings are a fraction of downloads and depend on how often an app prompts for one. A heavily prompted app can outrank a larger one.
• Resets happen. Apple can reset an app's ratings when a new version ships, so a drop between readings may be a reset, not a loss of users.
• Timing lags. The lookup interface can trail the store page by a day.
• Zeros are ambiguous. Many rows show 0 ratings, as Codictate does in the United Kingdom ⁴⁸. A zero means no ratings yet, not no users. The Mac store shows the same: Hold to Talk has 0 ⁴⁹, and Simon Says has 0 ⁵⁰.
• Windows and web apps are missing. Apps without a store listing are not in the data at all.
• Growth is not measured here. The question about speed needs the weekly series, and one reading is only a starting point.

Seller details also complicate comparisons. Siri's listing in the United Kingdom has 160 ratings ⁵¹ under a different Apple seller entity than in the United States, so even one product can be split across listings.

What this means if you are choosing how to write

Do not pick a dictation app by its rating count alone. A large count means an app is old, heavily prompted or bundled with a popular product. A small count on a recent app means little either way.

Read the average stars with the count. A count of a few hundred with a high average is a reasonable signal, while a handful of five-star ratings is not. Limitless has 125 ratings in the United States ⁵², but its average is far below most of its peers.

If you dictate in a language other than English, check your own storefront: an app popular in the United States may be nearly unrated in Japan or Brazil. For the basics of choosing between built-in tools and dedicated apps, read our overview of dictation software. For how these apps fit into the wider market, see our market page.

Free built-in dictation is enough for short messages. Dedicated apps earn their place for long documents, specialist vocabulary or work on a Mac. Rating counts can help you shortlist, but a short trial with your own voice is the better test.