Tracked by us ยท Dictation on phones and tablets
Dictation Apps on Google Play: Downloads and Ratings
How many times have dictation, voice keyboard and transcription apps been downloaded on Google Play, and how are they rated?
Google Play shows downloads in bands, from Gboard at 10,000,000,000 downloads down to young dictation apps with a few dozen, such as Contextli at 50 downloads. Big-tech keyboards, assistants and system tools dominate by volume, while standalone dictation apps such as Wispr Flow reach 1,000,000 downloads. Ratings are generally good, and the weakest scores sit on apps with few ratings or apps that ship preinstalled.
52 series, read every week from Google Play store pages (public)1 reading since October 19, 2026Updated October 19, 2026Free to reuse under CC BY 4.0
52data points
The google Play downloads of dictation and voice apps (tracked weekly) dataset holds 52 data points, each a published figure traced to its source.
How it is calculated
10,000,000,000downloads or more
Gboard showed 10B+ downloads on Google Play (at least 10,000,000,000), rated 4.5 stars from 18,894,235 ratings, on October 19, 2026.
How it is calculated
5,000,000,000downloads or more
Google Assistant showed 5B+ downloads on Google Play (at least 5,000,000,000), rated 3.8 stars from 1,365,031 ratings, on October 19, 2026.
How it is calculated
1,000,000downloads or more
Wispr Flow showed 1M+ downloads on Google Play (at least 1,000,000), rated 4.6 stars from 12,038 ratings, on October 19, 2026.
How it is calculated
Dictation on Android is mostly a story of platform software. The most downloaded voice input app in our registry is Gboard, which Google Play shows at 10,000,000,000 downloads 1, followed by Google Assistant at 5,000,000,000 downloads 2. Standalone dictation apps that people install on purpose sit several orders of magnitude lower: the leaders are around 1,000,000 downloads 3, and many sit in the thousands. The table below covers 52 apps 4, each read from its public US Google Play page on one day.
Keyboards and assistants reach billions, dictation apps reach millions
Google Play does not publish exact download counts. It publishes a band such as "1M+", and every number here is the floor of that band. A figure of 10,000,000,000 downloads 1 for Gboard means at least that many installs, not that many people.
The top of the table is crowded with products that come with a phone or sit inside a large platform:
| App | Band shown | Stars | Ratings | Downloads (lower bound of the band) (downloads) | Source |
|---|---|---|---|---|---|
| Gboard | 10B+ | 4.48 | 18894235 | 10,000,000,000 | Calculated 1 |
| Microsoft SwiftKey | 1B+ | 4.28 | 4599487 | 1,000,000,000 | Calculated 5 |
| FUTO Voice Input | 50K+ | 4.26 | 731 | 50,000 | Calculated 6 |
| Speakly | 10K+ | 4.18 | 69 | 10,000 | Calculated 7 |
| Dictate Keyboard | 5K+ | 4.58 | 699 | 5,000 | Calculated 8 |
| Offline Voice Input | 5K+ | 4.25 | 102 | 5,000 | Calculated 9 |
| Quill Flow | 5K+ | 4.25 | 982 | 5,000 | Calculated 10 |
Gboard and Microsoft SwiftKey (1,000,000,000 downloads 5) are general keyboards where voice typing is one button among many. Their download counts say how many phones carry a keyboard, not how many people dictate with it. The dedicated voice keyboards are far smaller. FUTO Voice Input shows 50,000 downloads 6, and Speakly shows 10,000 downloads 7. Their ratings are solid, but the audience is small.
Voice assistants are the second giant group, and AI chatbots joined it
The assistant segment holds the largest brand names:
| App | Band shown | Stars | Ratings | Downloads (lower bound of the band) (downloads) | Source |
|---|---|---|---|---|---|
| Google Assistant | 5B+ | 3.77 | 1365031 | 5,000,000,000 | Calculated 2 |
| ChatGPT voice mode | 1B+ | 4.78 | 65116475 | 1,000,000,000 | Calculated 11 |
| Gemini | 1B+ | 4.54 | 49314636 | 1,000,000,000 | Calculated 12 |
| Amazon Alexa | 100M+ | 4.06 | 5778064 | 100,000,000 | Calculated 13 |
| Grok | 100M+ | 4.87 | 4017818 | 100,000,000 | Calculated 14 |
| Meta AI | 100M+ | 4.62 | 2642446 | 100,000,000 | Calculated 15 |
| Perplexity | 100M+ | 4.63 | 2194598 | 100,000,000 | Calculated 16 |
| Claude voice mode | 50M+ | 4.51 | 807919 | 50,000,000 | Calculated 17 |
| Pi | 1M+ | 3.7 | 5537 | 1,000,000 | Calculated 18 |
Google Assistant leads. ChatGPT voice mode (1,000,000,000 downloads 11) and Gemini (1,000,000,000 downloads 12) have reached the same band as Android Voice Access (1,000,000,000 downloads 19). Grok, Meta AI, Perplexity and Amazon Alexa sit one band lower, at 100,000,000 downloads 14 for Grok, and Claude voice mode is lower again at 50,000,000 downloads 17.
These are chat and assistant apps first. Voice is one way to talk to them, so the download count includes many people who never speak to the app. They matter here because they compete for the same moment: you have something to say and a phone in your hand.
Standalone dictation apps split into a few leaders and a long tail
The dictation segment is where the spread is widest:
| App | Band shown | Stars | Ratings | Downloads (lower bound of the band) (downloads) | Source |
|---|---|---|---|---|---|
| Voice Notebook | 1M+ | 4.44 | 21157 | 1,000,000 | Calculated 20 |
| Wispr Flow | 1M+ | 4.57 | 12038 | 1,000,000 | Calculated 3 |
| Typeless | 500K+ | 4.16 | 3002 | 500,000 | Calculated 21 |
| Aqua Voice | 1K+ | 4.4 | 53 | 1,000 | Calculated 22 |
| Dictation Daddy | 1K+ | 3.88 | 17 | 1,000 | Calculated 23 |
| Superwhisper | 1K+ | 3.5 | 37 | 1,000 | Calculated 24 |
| Voicy | 1K+ | 4.29 | 12 | 1,000 | Calculated 25 |
| BossAI | 100+ | 100 | Calculated 26 | ||
| Contextli | 50+ | 50 | Calculated 27 |
Wispr Flow and Voice Notebook share the top band, at 1,000,000 downloads 3 and 1,000,000 downloads 20. Typeless follows at 500,000 downloads 21. Then comes a steep drop. Superwhisper and Aqua Voice both show 1,000 downloads 24, and BossAI shows 100 downloads 26. Contextli is at the bottom, with 50 downloads 27.
Two things explain the gap. First, several of these products began on desktop or iPhone, and their Android app is newer or secondary. Second, Android users often start with the free tools already on the phone, so a paid or account-based app has to give a clear reason to switch.
Transcription and meeting apps have a clear leader each
Transcription tools mix two kinds of products. Google's Live Transcribe is built into many phones and reaches 1,000,000,000 downloads 28. Paid and freemium services come next:
| App | Band shown | Stars | Ratings | Downloads (lower bound of the band) (downloads) | Source |
|---|---|---|---|---|---|
| Live Transcribe | 1B+ | 3.84 | 272588 | 1,000,000,000 | Calculated 28 |
| Speechnotes | 5M+ | 4.01 | 31743 | 5,000,000 | Calculated 29 |
| Transkriptor | 5M+ | 4.1 | 52142 | 5,000,000 | Calculated 30 |
| Notta | 1M+ | 4.18 | 21410 | 1,000,000 | Calculated 31 |
| Rev | 500K+ | 4.16 | 3164 | 500,000 | Calculated 32 |
| Riverside | 500K+ | 4.56 | 7799 | 500,000 | Calculated 33 |
| Good Tape | 10K+ | 10,000 | Calculated 34 | ||
| Recorder (Google Pixel) | 1+ | 3.23 | 15966 | 1 | Calculated 35 |
Speechnotes and Transkriptor each show 5,000,000 downloads 29, then Notta at 1,000,000 downloads 31. Rev and Riverside sit at 500,000 downloads 32. Good Tape is at 10,000 downloads 34.
Meeting notes follow the same shape on a smaller scale. Otter.ai leads with 5,000,000 downloads 36 and also carries one of the best ratings in the table. Fireflies.ai shows 500,000 downloads 37, Krisp 100,000 downloads 38, and Read AI 50,000 downloads 39. Wearable-linked or newer tools such as Limitless (5,000 downloads 40) and Bee (5,000 downloads 41) are far behind, and Limitless carries the lowest rating among the meeting tools.
Voice notes and professional tools are niches
Voice notes apps reach the hundred-thousands at best. Letterly and Voicenotes each show 100,000 downloads 42, AudioPen shows 50,000 downloads 43, and Flint shows 100 downloads 44. Letterly has the stronger rating of the first two.
Medical and professional dictation is smaller still. Nabla shows 5,000 downloads 45 and Lexacom 500 downloads 46. These products are sold to clinics and firms, so a small public install base fits how they are bought. Neither shows a star rating.
Ratings are high for most apps, but they rest on very different amounts of evidence
Most apps with a visible rating sit in the four-star range. The apps with the best scores include Grok, Otter.ai, Read AI, ChatGPT voice mode and Bluedot. The weakest are Limitless, Superwhisper, Pi and Google's own Recorder, followed by Android Voice Access, Google Assistant and Live Transcribe.
The pattern is useful. Tools bundled with a phone collect many ratings from people who never chose them, and those ratings run lower. Apps people install on purpose tend to score higher. But the count of ratings matters as much as the score. Dictation Daddy and Voicy have only a handful of ratings behind them, so their stars say almost nothing. Only a small share of installs ever leave a rating, so a score from a few dozen people should be read as a hint.
Where the data stops
- Bands, not counts. Growth only shows when an app crosses into the next band. An app can double and look unchanged.
- Installs, not users. A download is an install. It does not tell you how many people still use the app, or whether they use its voice features.
- Bundled apps inflate the top. Recorder (Google Pixel) shows 1 downloads 35 yet has a large number of ratings. It ships on Pixel phones, so its Play count misses nearly everyone who uses it. The same caution applies to Gboard, Live Transcribe and Android Voice Access.
- Missing ratings. Several apps, including Contextli, Good Tape, Nabla, Avoma, Leexi, Rimo and Bee, show no star rating on the day we read them, so they are left blank.
- Google Play only. Apps that live on iPhone, desktop or the web are outside this table. Several leading dictation tools are mostly used there, so low Android numbers do not mean low use overall.
- One day, US listing. Every row is the US page as shown on one date. Other countries and later dates can differ.
For wider context on who uses these tools and how the market fits together, see our market overview.
What this means if you are choosing how to write on a phone
Start with what your phone already has. Gboard and the built-in voice typing reach billions of people, and they cost nothing. If they handle your language and your accuracy needs, you do not need another app. Our guide to phones and tablets explains what built-in dictation can and cannot do.
Switch when you have a specific reason. If you need long notes, cleaned-up text, or recordings turned into transcripts, the transcription and voice notes groups hold the established choices, with Otter.ai at 5,000,000 downloads 36 and Transkriptor at 5,000,000 downloads 30. If you want offline, open-source voice input on a keyboard, FUTO Voice Input is the visible option.
Treat download bands as a signal of reach, not quality. A small app with a high rating from a few hundred people can fit you better than a giant with a middling score, and a large app with few ratings per install can still be mediocre. Try two or three candidates for a week on your own messages before you commit.
Evidence
Every number on this page, with where it comes from. Open a fact to see the exact words in its source.
- 1.
- 2.
- 3.
- 4.
- 5.
- 6.
- 7.
- 8.
- 9.
- 10.
- 11.
- 12.
- 13.
- 14.
- 15.
- 16.
- 17.
- 18.
- 19.
- 20.
- 21.
- 22.
- 23.
- 24.
- 25.
- 26.
- 27.
- 28.
- 29.
- 30.
- 31.
- 32.
- 33.
- 34.
- 35.
- 36.
- 37.
- 38.
- 39.
- 40.
- 41.
- 42.
- 43.
- 44.
- 45.
- 46.
Every data point
Each row is one published figure for one group, in downloads. The row opens its fact page with the exact words of the source; the source opens the original.
| Source | |||||||
|---|---|---|---|---|---|---|---|
| Voice Notebook | Voice Notebook | Dictation app | 1M+ | 4.44 | 21157 | 1,000,000 | Read October 19, 2026 |
| Wispr Flow | Wispr Flow | Dictation app | 1M+ | 4.57 | 12038 | 1,000,000 | Read October 19, 2026 |
| Typeless | Typeless | Dictation app | 500K+ | 4.16 | 3002 | 500,000 | Read October 19, 2026 |
| Aqua Voice | Aqua Voice | Dictation app | 1K+ | 4.4 | 53 | 1,000 | Read October 19, 2026 |
| Dictation Daddy | Dictation Daddy | Dictation app | 1K+ | 3.88 | 17 | 1,000 | Read October 19, 2026 |
| Superwhisper | Superwhisper | Dictation app | 1K+ | 3.5 | 37 | 1,000 | Read October 19, 2026 |
| Voicy | Voicy | Dictation app | 1K+ | 4.29 | 12 | 1,000 | Read October 19, 2026 |
| BossAI | BossAI | Dictation app | 100+ | 100 | Read October 19, 2026 | ||
| Contextli | Contextli | Dictation app | 50+ | 50 | Read October 19, 2026 | ||
| Gboard | Gboard | Voice keyboard | 10B+ | 4.48 | 18894235 | 10,000,000,000 | Read October 19, 2026 |
| Microsoft SwiftKey | Microsoft SwiftKey | Voice keyboard | 1B+ | 4.28 | 4599487 | 1,000,000,000 | Read October 19, 2026 |
| FUTO Voice Input | FUTO Voice Input | Voice keyboard | 50K+ | 4.26 | 731 | 50,000 | Read October 19, 2026 |
| Speakly | Speakly | Voice keyboard | 10K+ | 4.18 | 69 | 10,000 | Read October 19, 2026 |
| Dictate Keyboard | Dictate Keyboard | Voice keyboard | 5K+ | 4.58 | 699 | 5,000 | Read October 19, 2026 |
| Offline Voice Input | Offline Voice Input | Voice keyboard | 5K+ | 4.25 | 102 | 5,000 | Read October 19, 2026 |
| Quill Flow | Quill Flow | Voice keyboard | 5K+ | 4.25 | 982 | 5,000 | Read October 19, 2026 |
| Letterly | Letterly | Voice notes | 100K+ | 4.59 | 1523 | 100,000 | Read October 19, 2026 |
| Voicenotes | Voicenotes | Voice notes | 100K+ | 3.93 | 2686 | 100,000 | Read October 19, 2026 |
| AudioPen | AudioPen | Voice notes | 50K+ | 4.09 | 312 | 50,000 | Read October 19, 2026 |
| Audionotes | Audionotes | Voice notes | 10K+ | 4.11 | 409 | 10,000 | Read October 19, 2026 |
| Talknotes | Talknotes | Voice notes | 10K+ | 4.22 | 144 | 10,000 | Read October 19, 2026 |
| Flint | Flint | Voice notes | 100+ | 100 | Read October 19, 2026 | ||
| Live Transcribe | Live Transcribe | Transcription | 1B+ | 3.84 | 272588 | 1,000,000,000 | Read October 19, 2026 |
| Speechnotes | Speechnotes | Transcription | 5M+ | 4.01 | 31743 | 5,000,000 | Read October 19, 2026 |
| Transkriptor | Transkriptor | Transcription | 5M+ | 4.1 | 52142 | 5,000,000 | Read October 19, 2026 |
| Notta | Notta | Transcription | 1M+ | 4.18 | 21410 | 1,000,000 | Read October 19, 2026 |
| Rev | Rev | Transcription | 500K+ | 4.16 | 3164 | 500,000 | Read October 19, 2026 |
| Riverside | Riverside | Transcription | 500K+ | 4.56 | 7799 | 500,000 | Read October 19, 2026 |
| Good Tape | Good Tape | Transcription | 10K+ | 10,000 | Read October 19, 2026 | ||
| Recorder (Google Pixel) | Recorder (Google Pixel) | Transcription | 1+ | 3.23 | 15966 | 1 | Read October 19, 2026 |
| Otter.ai | Otter.ai | Meeting notes | 5M+ | 4.79 | 46270 | 5,000,000 | Read October 19, 2026 |
| Fireflies.ai | Fireflies.ai | Meeting notes | 500K+ | 4.63 | 7623 | 500,000 | Read October 19, 2026 |
| Krisp | Krisp | Meeting notes | 100K+ | 3.94 | 384 | 100,000 | Read October 19, 2026 |
| Read AI | Read AI | Meeting notes | 50K+ | 4.79 | 2460 | 50,000 | Read October 19, 2026 |
| Bluedot | Bluedot | Meeting notes | 10K+ | 4.69 | 163 | 10,000 | Read October 19, 2026 |
| Leexi | Leexi | Meeting notes | 10K+ | 10,000 | Read October 19, 2026 | ||
| Bee | Bee | Meeting notes | 5K+ | 5,000 | Read October 19, 2026 | ||
| Limitless | Limitless | Meeting notes | 5K+ | 2.93 | 73 | 5,000 | Read October 19, 2026 |
| Avoma | Avoma | Meeting notes | 1K+ | 1,000 | Read October 19, 2026 | ||
| Rimo | Rimo | Meeting notes | 1K+ | 1,000 | Read October 19, 2026 | ||
| Nabla | Nabla | Medical and professional | 5K+ | 5,000 | Read October 19, 2026 | ||
| Lexacom | Lexacom | Medical and professional | 500+ | 500 | Read October 19, 2026 | ||
| Android Voice Access | Android Voice Access | Built-in dictation | 1B+ | 3.86 | 246020 | 1,000,000,000 | Read October 19, 2026 |
| Google Assistant | Google Assistant | Voice assistant | 5B+ | 3.77 | 1365031 | 5,000,000,000 | Read October 19, 2026 |
| ChatGPT voice mode | ChatGPT voice mode | Voice assistant | 1B+ | 4.78 | 65116475 | 1,000,000,000 | Read October 19, 2026 |
| Gemini | Gemini | Voice assistant | 1B+ | 4.54 | 49314636 | 1,000,000,000 | Read October 19, 2026 |
| Amazon Alexa | Amazon Alexa | Voice assistant | 100M+ | 4.06 | 5778064 | 100,000,000 | Read October 19, 2026 |
| Grok | Grok | Voice assistant | 100M+ | 4.87 | 4017818 | 100,000,000 | Read October 19, 2026 |
| Meta AI | Meta AI | Voice assistant | 100M+ | 4.62 | 2642446 | 100,000,000 | Read October 19, 2026 |
| Perplexity | Perplexity | Voice assistant | 100M+ | 4.63 | 2194598 | 100,000,000 | Read October 19, 2026 |
| Claude voice mode | Claude voice mode | Voice assistant | 50M+ | 4.51 | 807919 | 50,000,000 | Read October 19, 2026 |
| Pi | Pi | Voice assistant | 1M+ | 3.7 | 5537 | 1,000,000 | Read October 19, 2026 |
52 rows. Open a row for the exact words of its source.
Calculated from the rows
Every statistic is calculated by code from the rows above. Open one to see the exact values it summarises.
How this dataset is built
- What counts as a row
- Every app in our market registry with a Google Play listing, as its public US page shows it on the day read.
- What does not
- Apps not on Google Play (iPhone-only, desktop and web apps).
- How rows are found and checked
- Read once a week from each app's public Google Play page: the downloads band and the structured rating data; every reading is kept.
- What the data cannot tell you
- Bands are wide (5M+ covers 5 to 10 million) and downloads are installs, not users.
Every reading
Each reading asks Google Play store pages (public) again and keeps the answers. The checksum identifies the archived answers of that day, so the published figures can be traced to them.
- October 19, 202652 of 53 requests answered, 52 figuressha256 592b8c561688af7d
The whole method, from search to calculation, is on the research data page.
Cite and reuse
You may copy, share and adapt this dataset for any purpose, including commercially, if you credit it with a link. Quoted passages remain their authors' words: cite the original source for those.
Yap (2026). Google Play downloads of dictation and voice apps (tracked weekly) [Data set]. https://yapapp.cc/data/play-installs
License: CC BY 4.0.
More datasets
- On-Device Speech Recognition Benchmark: 25 Languages56 data points
- Typing Speed Statistics: Average WPM by Group14 data points
- Speech vs Typing Speed: Text Entry Rates Compared82 data points
- Dictation Punctuation and Capitalization Accuracy56 data points
- Speaking Rate Statistics: Words per Minute15 data points
- Dictation in Noise: Speech Recognition Accuracy Tested15 data points
- Speech Recognition Accuracy: Word Error Rates641 data points
- Speech Recognition Speed: Real-Time Factor Data60 data points
- How Many Languages Speech Recognition Supports55 data points
- Speech-to-Text API Pricing: Cost per Hour of Audio34 data points
- Dictation Software Prices: What Each App Costs44 data points
- How Long Voice Assistants Keep Your Recordings40 data points
- Repetitive Strain Injury Statistics: How Common Is RSI36 data points
- Disability Statistics: Conditions That Make Typing Hard49 data points
- Voice Assistant and Voice Search Usage Statistics12 data points
- Dictation App Problems: What Users Complain About248 data points
- Speech Recognition Market Size: Every Estimate Compared9 data points
- Voice AI Funding: Speech and Dictation Startup Rounds34 data points
- History of Dictation Software: A Dated Timeline38 data points
- Dictation App Features Compared: Offline, Privacy, AI602 data points
- How Much Time People Spend Writing and Typing at Work15 data points
- Voice Recording Privacy Settlements and Fines13 data points
- Dictation App Store Ratings: How Many Each App Has540 data points
- Dictation App Ratings: Average Stars on the App Store326 data points
- Interest in Speech Recognition: Wikipedia Page Views39 data points
- Most Downloaded Speech Recognition Models (Hugging Face)263 data points
- Open-Source Speech Recognition on GitHub: Stars124 data points
- Speech-to-Text SDK Downloads: PyPI and npm102 data points
- Dictation on Hacker News: Mentions per Month23 data points
- Mac vs Windows Market Share: Desktop OS Data by Month36 data points
- Voice Typing Chrome Extensions: Users and Ratings20 data points
- Dictation Apps in the App Store Charts, by Country166 data points
- Speech Recognition Research Papers per Year10 data points
Read the overview: Dictation and Voice to Text on iPhone and Android