> ## Documentation Index
> Fetch the complete documentation index at: https://mintlify.com/facebookresearch/omnilingual-asr/llms.txt
> Use this file to discover all available pages before exploring further.

# Supported Languages

> Comprehensive guide to 1600+ supported languages and language ID format

Omnilingual ASR supports over **1,600 languages** across multiple scripts, including hundreds of languages never before covered by any ASR system.

## Language ID Format

Languages in Omnilingual ASR follow a standardized format:

```
{language_code}_{script}
```

### Format Components

<ParamField path="language_code" type="string" required>
  Three-letter ISO 639-3 language code (e.g., `eng` for English, `spa` for Spanish, `cmn` for Mandarin)
</ParamField>

<ParamField path="script" type="string" required>
  Four-letter ISO 15924 script code (e.g., `Latn` for Latin, `Arab` for Arabic, `Hans` for Simplified Chinese)
</ParamField>

### Examples

<CodeGroup>
  ```python Common Languages theme={null}
  "eng_Latn"  # English (Latin script)
  "spa_Latn"  # Spanish (Latin script)
  "fra_Latn"  # French (Latin script)
  "deu_Latn"  # German (Latin script)
  "cmn_Hans"  # Mandarin Chinese (Simplified)
  "cmn_Hant"  # Mandarin Chinese (Traditional)
  "jpn_Jpan"  # Japanese
  "kor_Hang"  # Korean (Hangul)
  "arb_Arab"  # Standard Arabic
  "rus_Cyrl"  # Russian (Cyrillic)
  "hin_Deva"  # Hindi (Devanagari)
  ```

  ```python Multiple Scripts theme={null}
  # Same language, different scripts
  "aze_Latn"  # Azerbaijani (Latin script)
  "aze_Cyrl"  # Azerbaijani (Cyrillic script)
  "aze_Arab"  # Azerbaijani (Arabic script)

  "kmr_Latn"  # Kurdish (Latin script)
  "kmr_Cyrl"  # Kurdish (Cyrillic script)
  "kmr_Arab"  # Kurdish (Arabic script)

  "urd_Arab"  # Urdu (Arabic script)
  "urd_Deva"  # Urdu (Devanagari script)
  "urd_Latn"  # Urdu (Latin script)
  ```

  ```python Regional Variants theme={null}
  # Dialect-specific variants
  "ell_Grek"             # Standard Greek
  "ell_Grek_cypr1249"    # Cypriot Greek

  "lld_Latn_gherd"       # Ladin (Gherdëina)
  "lld_Latn_valbadia"    # Ladin (Val Badia)

  "roh_Latn_surs1244"    # Romansh (Sursilvan)
  ```
</CodeGroup>

## Accessing the Language List

You can programmatically access the full list of supported languages:

<CodeGroup>
  ```python Basic Access theme={null}
  from omnilingual_asr.models.wav2vec2_llama.lang_ids import supported_langs

  # Print total count
  print(f"Total supported languages: {len(supported_langs)}")

  # Print all languages
  for lang in supported_langs:
      print(lang)
  ```

  ```python Check Specific Language theme={null}
  from omnilingual_asr.models.wav2vec2_llama.lang_ids import supported_langs

  # Check if a language is supported
  lang_id = "eng_Latn"
  if lang_id in supported_langs:
      print(f"{lang_id} is supported!")
  else:
      print(f"{lang_id} is not supported")
  ```

  ```python Filter by Script theme={null}
  from omnilingual_asr.models.wav2vec2_llama.lang_ids import supported_langs

  # Find all Latin-script languages
  latin_langs = [lang for lang in supported_langs if lang.endswith("_Latn")]
  print(f"Languages with Latin script: {len(latin_langs)}")

  # Find all Arabic-script languages
  arabic_langs = [lang for lang in supported_langs if lang.endswith("_Arab")]
  print(f"Languages with Arabic script: {len(arabic_langs)}")
  ```

  ```python Export to File theme={null}
  from omnilingual_asr.models.wav2vec2_llama.lang_ids import supported_langs
  import json

  # Export as JSON
  with open("supported_languages.json", "w") as f:
      json.dump(supported_langs, f, indent=2)

  # Export as CSV
  import csv
  with open("supported_languages.csv", "w", newline="") as f:
      writer = csv.writer(f)
      writer.writerow(["Language ID"])
      for lang in supported_langs:
          writer.writerow([lang])
  ```
</CodeGroup>

## Script Coverage

Omnilingual ASR supports multiple writing systems:

<Tabs>
  <Tab title="Latin Scripts">
    **Latin Script (Latn)** - Most widely used

    The majority of supported languages use Latin script, including:

    * European languages (English, Spanish, French, German, etc.)
    * African languages (Swahili, Yoruba, Hausa, etc.)
    * Southeast Asian languages (Indonesian, Vietnamese, Filipino, etc.)
    * Indigenous American languages (Quechua, Guarani, Nahuatl, etc.)

    ```python theme={null}
    "eng_Latn", "spa_Latn", "fra_Latn", "deu_Latn", "ita_Latn",
    "por_Latn", "pol_Latn", "nld_Latn", "ind_Latn", "vie_Latn"
    ```
  </Tab>

  <Tab title="Cyrillic Scripts">
    **Cyrillic Script (Cyrl)**

    Used primarily for Slavic and Central Asian languages:

    * Russian, Ukrainian, Belarusian
    * Serbian, Bulgarian, Macedonian
    * Kazakh, Kyrgyz, Tajik, Uzbek
    * Many minority languages of Russia

    ```python theme={null}
    "rus_Cyrl", "ukr_Cyrl", "bel_Cyrl", "bul_Cyrl", "srp_Cyrl",
    "mkd_Cyrl", "kaz_Cyrl", "mon_Cyrl", "tgk_Cyrl"
    ```
  </Tab>

  <Tab title="Arabic Scripts">
    **Arabic Script (Arab)**

    Used for Arabic and related languages:

    * Modern Standard Arabic and dialects
    * Persian (Farsi), Dari, Pashto
    * Urdu, Kurdish, Sindhi
    * Various Central Asian and African languages

    ```python theme={null}
    "arb_Arab", "fas_Arab", "urd_Arab", "pus_Arab", "ckb_Arab",
    "pbt_Arab", "azb_Arab", "kur_Arab"
    ```
  </Tab>

  <Tab title="Asian Scripts">
    **Asian Scripts**

    Multiple script families from Asia:

    * **Devanagari (Deva)**: Hindi, Nepali, Marathi, Sanskrit
    * **Bengali (Beng)**: Bengali, Assamese
    * **Tamil (Taml)**: Tamil
    * **Telugu (Telu)**: Telugu
    * **Kannada (Knda)**: Kannada
    * **Malayalam (Mlym)**: Malayalam
    * **Gujarati (Gujr)**: Gujarati
    * **Han (Hans/Hant)**: Chinese (Simplified/Traditional)
    * **Hangul (Hang)**: Korean
    * **Japanese (Jpan)**: Japanese
    * **Thai (Thai)**: Thai, Lao variants
    * **Myanmar (Mymr)**: Burmese, Mon
    * **Khmer (Khmr)**: Khmer, Cham variants

    ```python theme={null}
    "hin_Deva", "ben_Beng", "tam_Taml", "tel_Telu", "kan_Knda",
    "mal_Mlym", "guj_Gujr", "cmn_Hans", "cmn_Hant", "kor_Hang",
    "jpn_Jpan", "tha_Thai", "mya_Mymr", "khm_Khmr"
    ```
  </Tab>

  <Tab title="Other Scripts">
    **Additional Scripts**

    * **Ethiopic (Ethi)**: Amharic, Tigrinya
    * **Hebrew (Hebr)**: Hebrew, Yiddish
    * **Armenian (Armn)**: Armenian
    * **Georgian (Geor)**: Georgian, Svan
    * **Greek (Grek)**: Greek
    * **Tibetan (Tibt)**: Tibetan, Dzongkha
    * **Sinhala (Sinh)**: Sinhala
    * **Lao (Laoo)**: Lao
    * **Tifinagh (Tfng)**: Tamazight
    * **Canadian Syllabics (Cans)**: Cree, Ojibwe

    ```python theme={null}
    "amh_Ethi", "heb_Hebr", "hye_Armn", "kat_Geor", "ell_Grek",
    "bod_Tibt", "sin_Sinh", "lao_Laoo"
    ```
  </Tab>
</Tabs>

## Language Statistics

<CardGroup cols={3}>
  <Card title="Total Languages" icon="globe">
    **1,682** language-script combinations
  </Card>

  <Card title="High Performance" icon="chart-line">
    **78%** with CER below 10%
  </Card>

  <Card title="New Coverage" icon="star">
    **Hundreds** of previously uncovered languages
  </Card>
</CardGroup>

## Using Language IDs in Code

Here's how to use language IDs with the inference pipeline:

<CodeGroup>
  ```python Single Language theme={null}
  from omnilingual_asr.models.inference.pipeline import ASRInferencePipeline

  pipeline = ASRInferencePipeline(model_card="omniASR_LLM_Unlimited_7B_v2")

  # Transcribe English audio
  transcriptions = pipeline.transcribe(
      ["/path/to/english_audio.wav"],
      lang=["eng_Latn"],
      batch_size=1
  )
  ```

  ```python Multiple Languages theme={null}
  from omnilingual_asr.models.inference.pipeline import ASRInferencePipeline

  pipeline = ASRInferencePipeline(model_card="omniASR_LLM_Unlimited_7B_v2")

  audio_files = [
      "/path/to/english.wav",
      "/path/to/spanish.wav",
      "/path/to/mandarin.wav",
      "/path/to/arabic.wav"
  ]

  languages = [
      "eng_Latn",   # English
      "spa_Latn",   # Spanish
      "cmn_Hans",   # Mandarin (Simplified)
      "arb_Arab"    # Arabic
  ]

  transcriptions = pipeline.transcribe(
      audio_files,
      lang=languages,
      batch_size=4
  )
  ```

  ```python Validation Helper theme={null}
  from omnilingual_asr.models.wav2vec2_llama.lang_ids import supported_langs

  def validate_language_id(lang_id: str) -> bool:
      """Check if a language ID is supported."""
      if lang_id not in supported_langs:
          print(f"Warning: {lang_id} is not supported")
          print(f"Total supported languages: {len(supported_langs)}")
          return False
      return True

  # Usage
  lang = "eng_Latn"
  if validate_language_id(lang):
      print(f"{lang} is valid!")
  ```

  ```python Auto-detect Script theme={null}
  def get_languages_by_script(script: str) -> list:
      """Get all languages for a specific script."""
      from omnilingual_asr.models.wav2vec2_llama.lang_ids import supported_langs
      return [lang for lang in supported_langs if lang.endswith(f"_{script}")]

  # Get all Latin script languages
  latin_langs = get_languages_by_script("Latn")
  print(f"Found {len(latin_langs)} Latin-script languages")

  # Get all Cyrillic script languages
  cyrillic_langs = get_languages_by_script("Cyrl")
  print(f"Found {len(cyrillic_langs)} Cyrillic-script languages")
  ```
</CodeGroup>

## Performance by Language

For detailed per-language performance metrics, see the complete results in the [per\_language\_results\_table\_7B\_llm\_asr.csv](https://github.com/facebookresearch/omnilingual-asr/blob/main/per_language_results_table_7B_llm_asr.csv) file.

<Tip>
  The 7B LLM-ASR model achieves character error rates (CER) below 10% for 78% of supported languages.
</Tip>

## Language Families

Omnilingual ASR covers languages from diverse language families:

* **Indo-European**: Romance, Germanic, Slavic, Indo-Aryan, Iranian languages
* **Sino-Tibetan**: Chinese varieties, Tibetan, Burmese
* **Niger-Congo**: Bantu, Atlantic, Gur, Kwa languages
* **Austronesian**: Indonesian, Filipino, Polynesian languages
* **Afro-Asiatic**: Arabic, Amharic, Hausa, Hebrew
* **Dravidian**: Tamil, Telugu, Kannada, Malayalam
* **Turkic**: Turkish, Uzbek, Kazakh, Uyghur
* **Uralic**: Finnish, Hungarian, Estonian
* **Indigenous American**: Quechua, Guarani, Nahuatl, Aymara
* **And many more...**

## Finding Your Language

To find the correct language ID:

<Steps>
  <Step title="Identify ISO 639-3 Code">
    Look up your language's three-letter code at [ISO 639-3](https://iso639-3.sil.org/) or [Ethnologue](https://www.ethnologue.com/).

    Example: English = `eng`, Spanish = `spa`
  </Step>

  <Step title="Determine Script">
    Identify which script your text uses:

    * Latin alphabet → `Latn`
    * Cyrillic → `Cyrl`
    * Arabic → `Arab`
    * See [ISO 15924](https://unicode.org/iso15924/iso15924-codes.html) for complete list
  </Step>

  <Step title="Combine and Validate">
    Combine as `{language}_{script}` and check if it's in the supported list:

    ```python theme={null}
    from omnilingual_asr.models.wav2vec2_llama.lang_ids import supported_langs

    lang_id = "eng_Latn"
    if lang_id in supported_langs:
        print(f"✓ {lang_id} is supported")
    ```
  </Step>
</Steps>

## Complete Language List

The complete list of 1,682 supported languages is available in the source code at:

```
src/omnilingual_asr/models/wav2vec2_llama/lang_ids.py
```

A few example languages from the complete list:

<Accordion title="View Sample Languages (A-Z)">
  ```python theme={null}
  # Sample of supported languages
  "aae_Latn",  # Arbëreshë Albanian
  "aka_Latn",  # Akan
  "amh_Ethi",  # Amharic
  "arb_Arab",  # Standard Arabic
  "asm_Beng",  # Assamese
  "bam_Latn",  # Bambara
  "ben_Beng",  # Bengali
  "bul_Cyrl",  # Bulgarian
  "cat_Latn",  # Catalan
  "ces_Latn",  # Czech
  "cmn_Hans",  # Mandarin Chinese (Simplified)
  "cmn_Hant",  # Mandarin Chinese (Traditional)
  "dan_Latn",  # Danish
  "deu_Latn",  # German
  "ell_Grek",  # Greek
  "eng_Latn",  # English
  "epo_Latn",  # Esperanto
  "eus_Latn",  # Basque
  "fas_Arab",  # Persian
  "fin_Latn",  # Finnish
  "fra_Latn",  # French
  "gle_Latn",  # Irish
  "glg_Latn",  # Galician
  "grn_Latn",  # Guarani
  "guj_Gujr",  # Gujarati
  "hat_Latn",  # Haitian Creole
  "hau_Latn",  # Hausa
  "heb_Hebr",  # Hebrew
  "hin_Deva",  # Hindi
  "hrv_Latn",  # Croatian
  "hun_Latn",  # Hungarian
  "hye_Armn",  # Armenian
  "ibo_Latn",  # Igbo
  "ind_Latn",  # Indonesian
  "isl_Latn",  # Icelandic
  "ita_Latn",  # Italian
  "jav_Latn",  # Javanese
  "jpn_Jpan",  # Japanese
  "kan_Knda",  # Kannada
  "kat_Geor",  # Georgian
  "kaz_Cyrl",  # Kazakh
  "khm_Khmr",  # Khmer
  "kin_Latn",  # Kinyarwanda
  "kor_Hang",  # Korean
  "lao_Laoo",  # Lao
  "lat_Latn",  # Latin
  "lav_Latn",  # Latvian
  "lit_Latn",  # Lithuanian
  "lug_Latn",  # Luganda
  "mal_Mlym",  # Malayalam
  "mar_Deva",  # Marathi
  "mkd_Cyrl",  # Macedonian
  "mlg_Latn",  # Malagasy
  "mon_Cyrl",  # Mongolian
  "mri_Latn",  # Maori
  "mya_Mymr",  # Burmese
  "nep_Deva",  # Nepali
  "nld_Latn",  # Dutch
  "nob_Latn",  # Norwegian Bokmål
  "nya_Latn",  # Chichewa
  "ori_Orya",  # Odia
  "pan_Guru",  # Punjabi
  "pol_Latn",  # Polish
  "por_Latn",  # Portuguese
  "pus_Arab",  # Pashto
  "ron_Latn",  # Romanian
  "rus_Cyrl",  # Russian
  "sin_Sinh",  # Sinhala
  "slk_Latn",  # Slovak
  "slv_Latn",  # Slovenian
  "sna_Latn",  # Shona
  "som_Latn",  # Somali
  "spa_Latn",  # Spanish
  "sqi_Latn",  # Albanian
  "srp_Cyrl",  # Serbian
  "sun_Latn",  # Sundanese
  "swe_Latn",  # Swedish
  "swh_Latn",  # Swahili
  "tam_Taml",  # Tamil
  "tel_Telu",  # Telugu
  "tgk_Cyrl",  # Tajik
  "tha_Thai",  # Thai
  "tir_Ethi",  # Tigrinya
  "tuk_Latn",  # Turkmen
  "tur_Latn",  # Turkish
  "ukr_Cyrl",  # Ukrainian
  "urd_Arab",  # Urdu
  "uzb_Latn",  # Uzbek
  "vie_Latn",  # Vietnamese
  "wol_Latn",  # Wolof
  "xho_Latn",  # Xhosa
  "ydd_Hebr",  # Yiddish
  "yor_Latn",  # Yoruba
  "zho_Hans",  # Chinese (Simplified)
  "zul_Latn",  # Zulu
  # ... and 1,600+ more!
  ```
</Accordion>

## Next Steps

<CardGroup cols={2}>
  <Card title="Quick Start" icon="rocket" href="/quickstart">
    Start transcribing with language codes
  </Card>

  <Card title="Language Conditioning" icon="language" href="/advanced/language-conditioning">
    Improve accuracy with language hints
  </Card>

  <Card title="Model Selection" icon="sliders" href="/models/model-specifications">
    Choose the right model for your languages
  </Card>

  <Card title="Zero-Shot Learning" icon="wand-magic-sparkle" href="/models/zero-shot">
    Add new languages with examples
  </Card>
</CardGroup>