TED Talks are subtitled in over 100 languages, according to the TED languages page. The figures below show the number of talks covered by selected language editions, making subtitle availability a measurable part of how stories travel across media and geography.
Contents
- The scale of TED’s multilingual library
- The largest subtitle editions
- European language coverage
- Asian and Middle Eastern language coverage
- African, regional and smaller editions
- What the numbers show about access
The scale of TED’s multilingual library
The TED languages page reports subtitle coverage in a wide range of language editions. The top-level figure is “over 100 languages,” while the supplied language-by-language counts identify 58 editions. These are coverage counts for talks, not audience totals, view counts, speaker counts or translations of every item in the TED catalog. They also do not indicate when each subtitle record was added.
English is listed with 24,241 talks. Spanish follows with 18,337, while Arabic reaches 14,509. French has 11,949, and Brazilian Portuguese has 11,932. These figures illustrate the scale of multilingual access: several editions contain coverage in the five-figure range, while others contain only one or two talks.
The same source lists Portuguese, without the Brazilian regional label, at 8,321 talks. Chinese, Simplified is listed at 8,919, Chinese, Traditional at 7,584, and Chinese, Yue at 338. These labels should be read as separate entries in the TED language interface rather than combined into one Chinese total.
The figures are best understood as a snapshot associated with the TED languages page. Because no measurement date is supplied for the individual counts, they should not be treated as a dated historical series or as proof that the totals are unchanged today.
The largest subtitle editions
The largest listed coverage totals are concentrated among widely used international languages. English leads the supplied list at 24,241 talks. Spanish covers 18,337, and Arabic covers 14,509. French and French (Canada) are separate entries: French covers 11,949 talks, while French (Canada) covers 360.
The Portuguese entries also demonstrate why regional labels matter. Portuguese covers 8,321 talks, and Portuguese, Brazilian covers 11,932. The two displayed counts should not be added without a source-defined rule for overlap, because the figures may describe editions rather than mutually exclusive sets of talks.
| Language edition | Talks with subtitles |
|---|---|
| English | 24,241 |
| Spanish | 18,337 |
| Arabic | 14,509 |
| French | 11,949 |
| Portuguese, Brazilian | 11,932 |
| Italian | 9,089 |
| Chinese, Simplified | 8,919 |
| Vietnamese | 8,411 |
| Portuguese | 8,321 |
| Korean | 8,319 |
Italian is listed at 9,089 talks. Korean covers 8,319, Vietnamese 8,411, Russian 7,877, and Turkish 7,858. Japanese reaches 7,077, while Hebrew reaches 6,265. These numbers show substantial coverage beyond the largest English- and Spanish-language editions, with several language communities able to access thousands of subtitled talks.
Other five-thousand-plus editions include Romanian at 5,177 talks, Persian at 5,093, and Polish at 5,045. The distinction between these counts and total TED Talks is important: the figures measure talks with subtitles in a named edition, not the total number of talks available in the entire TED ecosystem.
European language coverage
The European entries span both large collections and more limited coverage. German is listed at 4,324 talks, Greek at 4,502, Dutch at 3,946, and Bulgarian at 2,686. Swedish covers 1,575 talks, while Hindi, although primarily associated with South Asia, is listed at 1,480 in the same set of language-page figures.
Several Central, Eastern and Southeastern European editions fall between several hundred and a few thousand talks. Croatian covers 2,385 talks, Czech 1,994, and Ukrainian 2,762. Kurdish (Central) is listed at 2,204. Hungarian covers 4,933, and Romanian 5,177.
The smaller European counts include Lithuanian at 880 talks, Albanian at 867, Catalan at 751, and Latvian at 721. Macedonian covers 441, Mongolian 433, Estonian 249, Belarusian 169, Esperanto 143, Azerbaijani 140, and Basque 106. These values show that subtitle access is not limited to only the largest language markets, even though the depth of coverage varies sharply.
The TED languages page also identifies Bosnian with 158 talks, Slovakian is not included in the supplied figures, and Serbian is not included either. Because the source list does not provide a complete catalog of every language edition, absence from this article is not evidence that a language lacks TED subtitles.
Danish is listed at 743 talks, and Dutch at 3,946. French has 11,949, while French (Canada) has 360. The regional French figure is a useful reminder that a language name and a regional edition can appear as distinct measurements in a subtitle system.
Asian and Middle Eastern language coverage
Asian language editions account for several of the largest counts after English and Spanish. Chinese, Simplified covers 8,919 talks, Chinese, Traditional 7,584, and Chinese, Yue 338. Japanese covers 7,077, Korean 8,319, and Vietnamese 8,411. Indonesian is listed at 4,552, Thai at 2,936, and Burmese at 2,465.
The figures also include Bengali at 215 talks, Assamese at 1, and Cebuano at 2. Hindi covers 1,480 talks. These counts should not be interpreted as a ranking of language importance or audience demand; they only describe the subtitle coverage reported for each named edition.
Persian covers 5,093 talks, Arabic 14,509, and Hebrew 6,265. Turkish is listed at 7,858, while Kurdish (Central) reaches 2,204. The spread from one talk in Assamese to 14,509 in Arabic is a reminder that multilingual publishing systems can contain very different levels of coverage within the same interface.
Other Asian and neighboring language figures include Thai at 2,936, Mongolian at 433, Armenian at 581, Azerbaijani at 140, and Ukrainian at 2,762. The source labels are retained here because regional and language distinctions matter when interpreting a subtitle count.
Vietnamese has 8,411 talks with subtitles, slightly above Portuguese at 8,321 and Korean at 8,319 in the supplied figures. Indonesian has 4,552, while Greek has 4,502. These close comparisons are descriptive only; they do not establish equivalence in audience reach, translation quality or publication rate.
African, regional and smaller editions
The supplied figures include several editions associated with Africa and regional language communities. Amharic covers 17 talks, Afrikaans 34, and Bislama 1. These are among the smallest listed counts. Arabic, by contrast, covers 14,509 talks, but the language-page figure alone does not identify how coverage is distributed across countries or dialect communities.
The list includes Algerian Arabic at 19 talks and Creole, Haitian at 8. These are separate labels from Arabic and from the other language editions. The counts therefore should not be combined into a single regional total without additional methodology.
There are also small entries for Asturian, with 2 talks, and Cebuano, with 2. Assamese and Bislama each cover 1 talk. Basque covers 106, while Catalan covers 751. Esperanto covers 143, and Bosnian covers 158. Such values can be useful indicators of the breadth of a platform’s volunteer or editorial translation footprint, but the dataset does not identify who produced the subtitles or how the work was coordinated.
The language page lists Romanian at 5,177, Polish at 5,045, and Russian at 7,877, alongside smaller entries such as Belarusian at 169 and Lithuanian at 880. This broad range makes the subtitle catalog a form of cross-media infrastructure: a single filmed talk can be made readable in multiple linguistic contexts, but the availability of that bridge depends on edition-specific coverage.
What the numbers show about access
The clearest pattern is uneven depth within broad reach. TED reports subtitles in over 100 languages, yet the supplied counts range from 1 talk to 24,241 talks. That means “available in a language” can describe very different practical experiences depending on the language selected.
At the high-coverage end, English has 24,241 talks and Spanish 18,337. Arabic has 14,509; French 11,949; Brazilian Portuguese 11,932; Italian 9,089; Chinese, Simplified 8,919; Vietnamese 8,411; Portuguese 8,321; and Korean 8,319. At the low-coverage end, Assamese and Bislama each have 1 talk, while Asturian and Cebuano each have 2.
Between those extremes is a long middle. German has 4,324 talks, Greek 4,502, Hungarian 4,933, Indonesian 4,552, and Dutch 3,946. Thai has 2,936, Ukrainian 2,762, Bulgarian 2,686, Burmese 2,465, and Croatian 2,385. The middle range is large enough to support meaningful browsing, but it remains much smaller than the largest editions.
These statistics describe subtitle coverage rather than the quality or usefulness of a translation. They also do not measure whether a talk’s visuals, examples or cultural references transfer successfully across languages. For storytelling across media, the numbers nevertheless identify a concrete distribution layer: subtitles determine how many recorded talks are potentially readable in each language edition.
The source is the TED languages page, and the figures are presented as legacy research rather than independently verified measurements. No date is supplied for the individual language counts, so readers should use them as reported coverage figures, not as a forecast, estimate of future growth or definitive current total.