Dataset · Languages and accents
How Many Languages Speech Recognition Supports
How many languages can each dictation feature, voice assistant, speech API and open model recognize?
45 data points from 39 sourcesUpdated October 19, 2026Free to reuse under CC BY 4.0
15languages
In the languages supported by speech recognition systems dataset, the 18 published figures for “Built-in dictation” have a median of 15 languages, ranging from 4 to 63.
How it is calculated
15languages
In the languages supported by speech recognition systems dataset, the 5 published figures for “Voice assistant” have a median of 15 languages, ranging from 14 to 45.
How it is calculated
94languages
In the languages supported by speech recognition systems dataset, the 10 published figures for “Cloud speech API” have a median of 94 languages, ranging from 13 to 125.
How it is calculated
62languages
In the languages supported by speech recognition systems dataset, the 10 published figures for “Open model” have a median of 62 languages, ranging from 13 to 1,600.
How it is calculated
Every data point
Each row is one published figure for one group, in languages. The row opens its fact page with the exact words of the source; the source opens the original.
| Source | ||||||||
|---|---|---|---|---|---|---|---|---|
| iOS Dictation (all variants) | iOS Dictation | Apple | Built-in dictation | Languages and regional variants | Not stated | 2026-10 | 63 | Calculated |
| Mac Dictation (all variants) | macOS Dictation | Apple | Built-in dictation | Languages and regional variants | Not stated | 2026-10 | 63 | Apple, 2026 |
| Safari speech recognition | Safari speech recognition (Web Speech API) | Apple | Built-in dictation | Not stated | 50 | WebKit (Apple), 2021 | ||
| iOS Dictation, on device | iOS Dictation, on-device and modeless | Apple | Built-in dictation | Languages and regional variants | Yes | 2026-10 | 48 | Apple, 2026 |
| Mac Dictation, on device | macOS Dictation, on-device and modeless | Apple | Built-in dictation | Languages and regional variants | Yes | 2026-10 | 48 | Apple, 2026 |
| Windows 11 voice typing | Windows 11 voice typing | Microsoft | Built-in dictation | Languages and regional variants | No | 2026-10 | 46 | Microsoft, 2026 |
| Claude Code voice dictation | Claude Code voice dictation | Anthropic | Built-in dictation | Languages | Not stated | 2026-10 | 20 | Anthropic, 2026 |
| Samsung Transcript assist, US | Samsung Transcript assist (US) | Samsung | Built-in dictation | Languages | Not stated | 16 | Samsung Support (US) | |
| Windows Voice Access | Windows Voice Access | Microsoft | Built-in dictation | Languages and regional variants | Yes | 2026-10 | 15 | Microsoft Support |
| Dictate in Word, main list | Dictate in Word | Microsoft | Built-in dictation | Languages and regional variants | Not stated | 15 | Microsoft, 2026 | |
| Samsung Transcript assist, UAE | Samsung Transcript assist (UAE) | Samsung | Built-in dictation | Languages | Not stated | 13 | Samsung Support (UAE), 2026 | |
| Windows 10 dictation | Windows 10 dictation (Windows key + H) | Microsoft | Built-in dictation | Languages and regional variants | Not stated | 13 | Microsoft | |
| Claude mobile app dictation | Claude mobile app dictation | Anthropic | Built-in dictation | Languages | Not stated | 2026-10 | 12 | Anthropic, 2026 |
| Windows Speech Recognition | Windows Speech Recognition | Microsoft | Built-in dictation | Languages and regional variants | Not stated | 11 | Microsoft, 2026 | |
| Google Assistant voice typing | Google Assistant voice typing | Built-in dictation | Languages | Not stated | 2026-10 | 6 | ||
| Gboard advanced voice typing | Gboard advanced voice typing | Built-in dictation | Languages | Not stated | 2026-10 | 6 | Google, 2026 | |
| Windows Voice Access, command list | Windows 11 Voice Access (command list) | Microsoft | Built-in dictation | Languages | Not stated | 6 | Microsoft Support, 2026 | |
| Windows Voice Access, FAQ | Windows 11 Voice Access (FAQ) | Microsoft | Built-in dictation | Languages | Not stated | 4 | Microsoft, 2026 | |
| Gemini in Android Auto | Gemini in Android Auto | Voice assistant | Languages | Not stated | 2026-10 | 45 | Google, The Keyword blog, 2025 | |
| Mac Voice Control | macOS Voice Control | Apple | Voice assistant | Languages and regional variants | Not stated | 2026-10 | 20 | Apple, 2026 |
| Alexa, AVS locales | Alexa Voice Service (Alexa Built-in) | Amazon | Voice assistant | Languages and regional variants | Not stated | 2026-10 | 15 | Amazon |
| Google Assistant Service API | Google Assistant Service API | Voice assistant | Languages and regional variants | Not stated | 2026-10 | 14 | Google Developers, 2024 | |
| Google Assistant on Google TV | Google Assistant on Google TV | Voice assistant | Languages | Not stated | 2026-10 | 14 | ||
| Google Cloud Speech-to-Text | Google Cloud Speech-to-Text | Cloud speech API | Languages | Not stated | 2026-10 | 125 | Cloud, 2026 | |
| Amazon Transcribe, key features | Amazon Transcribe | Amazon Web Services | Cloud speech API | Not stated | 2026-10 | 100 | Services | |
| Azure AI Speech, captioning | Microsoft Azure AI Speech | Microsoft | Cloud speech API | Not stated | 2026-10 | 100 | Microsoft | |
| AssemblyAI Universal-2 | AssemblyAI Universal-2 | AssemblyAI | Cloud speech API | Languages | Not stated | 2026-10 | 99 | AssemblyAI, 2026 |
| OpenAI Whisper API | Whisper (OpenAI Transcriptions API) | OpenAI | Cloud speech API | Languages | No | 98 | OpenAI, 2026 | |
| ElevenLabs Scribe v2 | ElevenLabs Scribe v2 | ElevenLabs | Cloud speech API | Languages | Not stated | 2026-10 | 90 | ElevenLabs, 2026 |
| Deepgram Nova-3 | Deepgram Nova-3 | Deepgram | Cloud speech API | Languages | Not stated | 2026-10 | 60 | Deepgram |
| Speechmatics STT (Enhanced, Standard) | Speechmatics transcription (Enhanced and Standard) | Speechmatics | Cloud speech API | Languages | Not stated | 2026-10 | 55 | Speechmatics, 2026 |
| AssemblyAI Universal-3.5 Pro | AssemblyAI Universal-3.5 Pro | AssemblyAI | Cloud speech API | Languages | Not stated | 2026-10 | 18 | AssemblyAI, 2026 |
| Voxtral Mini Transcribe V2 | Voxtral Mini Transcribe V2 | Mistral AI | Cloud speech API | Languages | Not stated | 2026-10 | 13 | AI, 2026 |
| Omnilingual ASR 7B | Omnilingual ASR (omniASR-LLM-7B) | Meta | Open model | Languages | Not stated | 2026-10 | 1,600 | FAIR, 2025 |
| Meta MMS, mms-1b-all | Meta MMS mms-1b-all checkpoint | Meta | Open model | Languages | Not stated | 2026-10 | 1,162 | Meta, 2023 |
| Meta MMS ASR | Meta MMS multilingual ASR model | Meta | Open model | Languages | Not stated | 2026-10 | 1,107 | Pratap et al., 2023 |
| Meta SeamlessM4T | SeamlessM4T | Meta | Open model | Languages | Not stated | 100 | Communication, 2023 | |
| Radford et al. 2022, Whisper | Whisper (model paper) | OpenAI | Open model | Languages | Not stated | 99 | Radford et al., 2022 | |
| NVIDIA Canary 1B v2 | NVIDIA Canary 1B v2 | NVIDIA | Open model | Languages | Not stated | 2026-10 | 25 | NVIDIA, 2025 |
| NVIDIA Parakeet TDT 0.6B v3 | NVIDIA Parakeet TDT 0.6B v3 | NVIDIA | Open model | Languages | Not stated | 2026-10 | 25 | NVIDIA, 2025 |
| IndicASR | IndicASR | AI4Bharat, IIT Madras | Open model | Languages | Not stated | 2026-10 | 22 | Javed et al., 2024 |
| Vosk | Vosk | Alpha Cephei | Open model | Yes | 20 | Alpha Cephei, 2026 | ||
| Voxtral Realtime | Voxtral Realtime | Mistral AI | Open model | Languages | Yes | 2026-10 | 13 | AI, 2026 |
| Heidi Dictate | Heidi Dictate | Heidi Health | Dictation app | Languages | Not stated | 2026-10 | 90 | Heidi Health |
| Dragon Pro Individual 15.6.1 | Dragon Professional Individual 15.6.1 | Nuance Communications | Dictation app | Languages | Not stated | 2026-10 | 6 | Communications |
45 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.
- 15 languagesIn the languages supported by speech recognition systems dataset, the 18 published figures for “Built-in dictation” have a median of 15 languages, ranging from 4 to 63.Calculated from other facts
- 15 languagesIn the languages supported by speech recognition systems dataset, the 5 published figures for “Voice assistant” have a median of 15 languages, ranging from 14 to 45.Calculated from other facts
- 94 languagesIn the languages supported by speech recognition systems dataset, the 10 published figures for “Cloud speech API” have a median of 94 languages, ranging from 13 to 125.Calculated from other facts
- 62 languagesIn the languages supported by speech recognition systems dataset, the 10 published figures for “Open model” have a median of 62 languages, ranging from 13 to 1,600.Calculated from other facts
- 48 languagesIn the languages supported by speech recognition systems dataset, the 2 published figures for “Dictation app” have a median of 48 languages, ranging from 6 to 90.Calculated from other facts
- 45 data pointsThe languages supported by speech recognition systems dataset holds 45 data points, each a published figure traced to its source.Calculated from other facts
- 39 sourcesThe languages supported by speech recognition systems dataset draws on 39 distinct sources.Calculated from other facts
How this dataset is built
- What counts as a row
- The number of languages (or language and region variants) a named speech recognition system officially supports, from its maker's documentation or model paper, dated.
- What does not
- Counts from third-party listicles, and translation languages that are not recognized from speech.
- How rows are found and checked
- Rows come from the facts already cited across the library and from each maker's current documentation, read on the date shown. Where a maker counts regional variants (English (US), English (UK)) the row says so.
- What the data cannot tell you
- Makers count differently: some list languages, others locales. Support also varies by device and feature, so a language on the list may not work offline or everywhere.
The search record
93
works found by the searches below
50
report a figure in the dataset's terms
20
read and judged by a researcher
18
kept, with their figures checked
The 4 searches, last run October 19, 2026
- OpenAlex: “multilingual speech recognition number of languages model”, 50 results
- arXiv: “multilingual speech recognition number of languages model”, 0 results
- OpenAlex: “massively multilingual speech recognition languages supported”, 50 results
- arXiv: “massively multilingual speech recognition languages supported”, 9 results
Every work and page screened (20: 18 kept)
Kept, 4 figuresApple Dictation on Mac and iPhone
Apple's iOS and macOS feature availability pages list Dictation languages by language and region, with a separate on-device list; the rows count those entries (63 general, 48 on-device on each system), since Apple prints no total. Rechecked 2026-10-19.
Not keptSiri
Apple's 'Change Siri voice or language' support page and the iOS feature availability page give no count of languages Siri recognizes: the page says languages vary by region and device, and its Siri lists are per country and per Apple Intelligence feature, so no total is stated.
Kept, 1 figureWindows voice typing
Microsoft's Use voice typing article (Windows 11) lists 46 language and locale entries that support voice typing; the existing fact is reused and its quote was rechecked against the live page on 2026-10-19.
Kept, 1 figureWindows Voice Access
Microsoft's Voice access setup page lists 15 languages and dialects (six English, two Spanish, two French, two Chinese, German, Japanese, Italian). Its FAQ names four languages and its command list six, and a Windows deprecation note says five English locales while listing six; those three disagree with the setup page, so only the setup page is counted.
Not keptGboard voice typing
No official Google page gives a count of languages for Gboard voice typing. The six-language list on the advanced voice typing help page covers voice commands for writing tools, not dictation, so it is not counted. A 2017 Google blog figure of 119 language varieties is dated and covers speech recognition across Google products, not Gboard voice typing today.
Kept, 1 figureGoogle Cloud Speech-to-Text
Google's Speech-to-Text product page states it supports over 125 languages, a floor, so 125 is the row. The same page gives 85+ languages and variants in its Chirp 3 transcription description; that is a different counting basis and not a separate row here.
Kept, 1 figureMicrosoft Azure AI Speech
Product page says audio captioning works in more than 100 languages, recorded as a floor of 100. The Learn locale table lists 148 locales, but it states no total and cannot be quoted, so that count is not recorded.
Kept, 1 figureAmazon Transcribe
Product page says key features work across 100+ languages, recorded as a floor of 100. The supported languages table lists 113 language and locale entries, but it states no total and cannot be quoted, so that count is not recorded.
Kept, 1 figureOpenAI Whisper (open model)
Model paper says the training set covers 99 languages, one language token each. Reuses the published fact. The 98 in OpenAI's API guide describes the hosted whisper-1 model, which is a separate row.
Kept, 2 figuresMeta MMS (open model)
Two official language counts: 1162 on the mms-1b-all model card (2026-10 read) and 1,107 in the 2023 MMS paper; the card's usage example also says 1107, and the paper is kept as its own row.
Kept, 1 figureNVIDIA Parakeet TDT 0.6B v3 (open model)
NVIDIA's model card states 25 European languages for the model (2026-10 read); reused existing fact, and its Supported Languages list gives the same languages by code.
Kept, 1 figureNVIDIA Canary 1B v2 (open model)
NVIDIA's model card states speech transcription for 25 languages (2026-10 read); reused existing fact. Its 24-language English-to-X translation count is not a recognition row.
Kept, 1 figureDeepgram
Official language table gives a Nova-3 language list with no stated total; we count 60 languages from it, and the count is noted as ours.
Kept, 2 figuresAssemblyAI
Official supported-languages page states totals: Universal-2 supports 99 languages and Universal-3.5 Pro supports 18.
Kept, 1 figureElevenLabs Scribe
Official speech-to-text documentation states Scribe v2 supports 90+ languages, a floor and not an exact total.
Kept, 1 figureSpeechmatics
Speechmatics' own Languages page lists its transcription languages; the row counts 55 single languages from the table (not the auto option or the seven bilingual packs), since the page prints no total. Read 2026-10-19.
Kept, 1 figureDragon
Nuance's 15.6 release notes state that the Dragon Professional Individual 15.6.1 full build supports six languages (ENX, DEU, FRA, ITA, ESP, NLD). No official list for Dragon 16 was found; the Nuance feature matrix PDF has none. The Dragonfly page (a third-party source for version 15) is not used. Read 2026-10-19.
Kept, 1 figureAmazon Alexa
Amazon's Alexa Voice Service developer page lists 15 language and region entries for Alexa, covering nine languages; this is the device-maker list, not the consumer app's language menu, which could not be read. Read 2026-10-19.
Kept, 3 figuresGoogle Assistant
Google's own pages state language counts: the Assistant Service developer table (14 codes, 8 languages), Assistant on Google TV (14 languages), and voice typing (six languages, already in the library). The Nest language page gives no total and says both 2 and 3 languages, so it is not counted.
Kept, 2 figuresMistral Voxtral
Mistral's Voxtral Transcribe 2 announcement states 13 languages for Voxtral Mini Transcribe V2 and for Voxtral Realtime. The 2025 Voxtral paper and the Mistral docs pages give no language count in the text read, so no row for the 2025 models.
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). Languages supported by speech recognition systems [Data set]. https://yapapp.cc/data/asr-language-support
License: CC BY 4.0.
Pages built on this data
- Google Assistant Voice Commands and What Gemini Changes
- Why are voice commands not working on Android Auto?
- How many languages does iPhone dictation support?
- Can Siri, Alexa or Gemini take long dictation?
- What voice commands does Google Assistant understand?
- Open-Source Speech Recognition: The Best Models
- Dictation in ChatGPT and Claude: Speaking Your Prompts
- Voice Input for Claude Code, Cursor and AI Coding Tools
- What languages does Claude dictation support?
- Why is Claude dictation not working?
- Can speech to text handle Hinglish, Spanglish and other mixed languages?
- Can you dictate in two languages at the same time?
- Does dictation detect the language automatically?
- How do you change the voice typing language on Android?
- How do you dictate in Japanese on a phone or computer?
- Dictation Languages, Accents and Voice Translation
- Dictation Languages: What Each App and Device Supports
- Speech Recognition in Python: Libraries Compared
- How can you dictate in a language your device does not support?
- How many languages does Google Speech-to-Text support?
- Which dictation app supports the most languages?
- What is the Heidi dictation app?
- Hindi Voice Typing and Speech to Text
- Which speech to text API works best for Indian languages?
More datasets
- On-Device Speech Recognition Benchmark on a Mac26 data points
- Typing Speed Statistics: Average WPM by Group14 data points
- Speech vs Typing Speed: Text Entry Rates Compared57 data points
- Speaking Rate Statistics: Words per Minute7 data points
- Speech Recognition Accuracy: Word Error Rates597 data points
- Speech Recognition Speed: Real-Time Factor Data24 data points
- Speech-to-Text API Pricing: Cost per Hour of Audio34 data points
- Dictation Software Prices: What Each App Costs37 data points
- How Long Voice Assistants Keep Your Recordings38 data points
- Repetitive Strain Injury Statistics: How Common Is RSI18 data points
- Disability Statistics: Conditions That Make Typing Hard22 data points
- Voice Assistant and Voice Search Usage Statistics2 data points
Read the overview: Dictation Languages, Accents and Voice Translation