Updating Azure STT and Translation language maps.

python/AzureSpeechDetection/acs_speech_component/azure_utils.py

@@ -28,140 +28,159 @@
 #  BCP-47 code in each value list will be used (the rest are primarily for
 #  reference in case of later changes or customization)
 
+# This mapping is also used to generate `ISO_LANGUAGE` codes after
+# Speech-to-Text conversion. The last ISO code listed in the key list will be used.
+
 # Supported languages can be found here:
 # https://learn.microsoft.com/en-us/azure/ai-services/speech-service/language-support?tabs=stt
 ISO6393_TO_BCP47 = dict(
-    afr=["af-ZA"],
-    amh=["am-ET"],
+    afr=["af-ZA"], # Afrikaans
+    amh=["am-ET"], # Amharic
     ara=["ar-EG", "ar-SA", "ar-IQ", "ar-IL", "ar-AE", "ar-SY", "ar-LY", "ar-DZ",
          "ar-BH", "ar-JO", "ar-KW", "ar-LB", "ar-MA", "ar-OM", "ar-PS", "ar-QA",
-         "ar-TN", "ar-YE"],
-    aze=["az-AZ"],
+         "ar-TN", "ar-YE"], # Arabic
     azj=["az-AZ"], # North Azerbaijani
     azb=["az-AZ"], # South Azerbaijani
+    aze=["az-AZ"], # Azerbaijani (Inclusive)
     # bel=["be-BY"], # Deprecated
     ben=["bn-IN"], # "bn-BD" Bengali-Bangladesh has been deprecated
     # bod=["bo"], # Deprecated
-    bul=["bg-BG"],
-    bos=["bs-BA"],
-    cat=["ca-ES"],
+    bul=["bg-BG"], # Bulgarian
+    bos=["bs-BA"], # Bosnian
+    cat=["ca-ES"], # Catalan, Valencian
     # ceb=["ceb"],
-    ces=["cs-CZ"],
-    cze=["cs-CZ"],# ISO-639-2 Variant
-    cym=["cy-GB"],
-    wel=["cy-GB"],# ISO-639-2 Variant
+    cze=["cs-CZ"], # Czech SO-639-2 Variant
+    ces=["cs-CZ"], # Czech
+    wel=["cy-GB"], # Welsh ISO-639-2 Variant
+    cym=["cy-GB"], # Welsh
     dan=["da-DK"], # Note: There is a related dialect JUT - Jutlandic
     jut=["da-DK"], # Upon further research, Jutlantic is present in Denmark
                    # but declining over time.
-    deu=["de-DE", "de-AT", "de-CH"],
+    ger=["de-DE", "de-AT", "de-CH"], # German ISO-2 Variant
+    deu=["de-DE", "de-AT", "de-CH"], # German
     # Many other forms of German exist
     gsw=["de-CH"], # Swiss German
     bar=["de-AT"], # Bavarian / Upper German variant common in most of Austria
-    ell=["el-GR"],
+    gre=["el-GR"], # Greek ISO-639-2 Variant
+    ell=["el-GR"], # Greek
     eng=["en-US", "en-CA", "en-GB", "en-AU", "en-GH", "en-HK", "en-IN", "en-IE",
          "en-KE", "en-NZ", "en-NG", "en-PH", "en-SG", "en-ZA", "en-TZ"],
-    est=["et-EE"], # Estonian (Inclusive)
     ekk=["et-EE"], # Standard Estonian
+    est=["et-EE"], # Estonian (Inclusive)
     # vro=["et-EE"], Voro, doesn't seem to be direct match
-    eus=["eu-ES"],
-    fas=["fa-IR"],
-    fin=["fi-FI"],
-    fil=["fil-PH"],
-    fra=["fr-FR", "fr-BE", "fr-CA", "fr-CH"],
-    gle=["ga-IE"],
-    glg=["gl-ES"],
-    guj=["gu-IN"],
-    heb=["he-IL"],
-    hin=["hi-IN"],
-    hrv=["hr-HR"],
-    hun=["hu-HU"],
+    baq=["eu-ES"], # Basque ISO-639-2 Variant
+    eus=["eu-ES"], # Basque
+    fin=["fi-FI"], # Finnish
+    fre=["fr-FR", "fr-BE", "fr-CA", "fr-CH"], # French ISO-639-2 Variant
+    fra=["fr-FR", "fr-BE", "fr-CA", "fr-CH"], # French
+    gle=["ga-IE"], # Irish
+    glg=["gl-ES"], # Galician
+    guj=["gu-IN"], # Gujarati
+    heb=["he-IL"], # Hebrew
+    hin=["hi-IN"], # Hindi
+    hrv=["hr-HR"], # Croatian
+    hun=["hu-HU"], # Hungarian
     # ohu=["hu-HU"], # Note: Old-Hungarian, might not fully work with modern "hu-HU"
     # gug=["gn"], # Deprecated
     # hat=[],
     # hau=["ha"], # Deprecated
     # hbs=["sh"], # Deprecated
-    # hye=["hy"],
-    ita=["it-IT", "it-CH"],
-    ind=["id-ID"],
-    ice=["is-IS"],
-    isl=["is-IS"],
-    jav=["jv-ID"],
-    jpn=["ja-JP"],
-    kat=["ka-GE"],
-    kaz=["kk-KZ"],
-    khm=["km-KH"],
+    arm=['hy-AM'], # Armenian ISO-639-2 Variant
+    hye=['hy-AM'], # Armenian
+    ita=["it-IT", "it-CH"], # Italian
+    ind=["id-ID"], # Indonesian
+    ice=["is-IS"], # Icelandic ISO-639-2 Variant
+    isl=["is-IS"], # Icelandic
+    jav=["jv-ID"], # Javanese
+    jpn=["ja-JP"], # Japanese
+    geo=["ka-GE"], # Georgian ISO-639-2 Variant
+    kat=["ka-GE"], # Georgian
+    kaz=["kk-KZ"], # Kazakh
     kxm=["km-KH"], # Northern Khmer, might not work as well.
-    kan=["kn-IN"],
+    khm=["km-KH"], # Central Khmer
+    kan=["kn-IN"], # Kannada
     # kir=["ky-KG"], # Deprecated
-    kor=["ko-KR"],
+    kor=["ko-KR"], # Korean
     # kur=["ku"], # Deprecated
-    lao=["lo-LA"],
-    lit=["lt-LT"],
-    lav=["lv-LV"],
+    lao=["lo-LA"], # Lao
+    lit=["lt-LT"], # Lithuanian
     lvs=["lv-LV"], # Standard Latvian
+    lav=["lv-LV"], # Latvian (Inclusive)
     # luo=[],
-    mkd=["mk-MK"],
-    mya=["my-MM"],
-    mal=["ml-IN"],
-    mon=["mn-MN"], # Mongolian (Inclusive)
+    mac=["mk-MK"], # Macedonian ISO 639-2 Variant
+    mkd=["mk-MK"], # Macedonian
+    bur=["my-MM"], # Burmese / Myanmar ISO 639-2 Variant
+    mya=["my-MM"], # Burmese / Myanmar
+    mal=["ml-IN"], # Malayalam
     khk=["mn-MN"], # Khalkha Mongolian (Predominant)
     mvf=["mn-MN"], # Peripheral Mongolian (Part)
-    mar=["mr-IN"],
-    zsm=["ms-MY"],
-    mlt=["mt-MT"],
-    nob=["nb-NO"],
-    nep=["ne-NP"], # Nepali (Macrolanguage)
+    mon=["mn-MN"], # Mongolian (Inclusive)
+    mar=["mr-IN"], # Marathi
+    may=["ms-MY"], # Malay ISO 639-2 Variant
+    msa=["ms-MY"], # Malay (Inclusive, Macrolanguage)
+    zsm=["ms-MY"], # Standard Malay (Malaysian Malay)
+    # In this case, the ms-MY code indicates ZSM (Standard Malay)
+    mlt=["mt-MT"], # Maltese
+    nob=["nb-NO"], # Norwegian Bokmål (Norway)
     npi=["ne-NP"], # Nepali
-    nld=["nl-NL", "nl-BE"], # Netherlands and Belgium
+    nep=["ne-NP"], # Nepali (Inclusive, Macrolanguage)
+    dut=["nl-NL", "nl-BE"], # Dutch ISO 639-2 Variant
+    nld=["nl-NL", "nl-BE"], # Dutch - Netherlands and Belgium
     # omr=["mr-IN"], # Old Maranthi, might not work
     # nde=["nd"],
     # orm=["om"],
-    pan=["pa-IN"],
-    pes=["fa-IR"],
-    pol=["pl-PL"],
+    pan=["pa-IN"], # Punjabi, Panjabi
+    per=["fa-IR"], # Persian ISO-639-2 Variant
+    fas=["fa-IR"], # Persian
+    pes=["fa-IR"], # Iranian Persian
+    pol=["pl-PL"], # Polish
     por=["pt-BR", "pt-PT"], # pt-BR = Portuguese Brazil, pt-PT = Portuguese Portugal
-    pus=["ps-AF"], # Pashto, Pushto (Inclusive)
     pbu=["ps-AF"], # Northern Pahsto
     pst=["ps-AF"], # Central Pahsto
     pbt=["ps-AF"], # Southern Pahsto
-    sin=["si-LK"],
+    pus=["ps-AF"], # Pashto, Pushto (Inclusive)
+    sin=["si-LK"], # Sinhala, Sinhalese
     # prs=["prs-AF"], # Deprecated
     # pus=["pa-AF"], # Deprecated
-    ron=["ro-RO"],  # ro-MD deprecated
+    rum=["ro-RO"], # Romanian ISO-639-2 Variant
+    ron=["ro-RO"], # Romanian, Moldavian, Moldovan
+    # ro-MD deprecated
     # run=[],
-    rus=["ru-RU"],
-    slk=["sk-SK"],
-    slv=["sl-SI"],
+    rus=["ru-RU"], # Russian
+    slo=["sk-SK"], # Slovak ISO-639-2 Variant
+    slk=["sk-SK"], # Slovak
+    slv=["sl-SI"], # Slovenian
     # sna=["sn"],
-    som=["so-SO"],
+    som=["so-SO"], # Somali
     spa=["es-MX", "es-US", "es-AR", "es-BO", "es-CL", "es-CO", "es-CR", "es-CU",
          "es-DO", "es-EC", "es-SV", "es-GQ", "es-GT", "es-HN", "es-NI", "es-PA",
-         "es-PY", "es-PE", "es-PR", "es-ES", "es-UY", "es-VE"],
-
-    sqi=["sq-AL"],
-    swa=["sw-KE", "sw-TZ"],
-    swe=["sv-SE"],
-    srp=["sr-RS"],
-    tam=["ta-IN"],
-    tel=["te-IN"],
+         "es-PY", "es-PE", "es-PR", "es-ES", "es-UY", "es-VE"], # Spanish
+    alb=["sq-AL"], # Albanian ISO-639-2 Variant
+    sqi=["sq-AL"], # Albanian
+    swa=["sw-KE", "sw-TZ"], # Swahili
+    swe=["sv-SE"], # Swedish
+    srp=["sr-RS"], # Serbian
+    tam=["ta-IN"], # Tamil
+    tel=["te-IN"], # Telugu
     # wbq = ["te-IN"], Waddar/Vadari is related to Telugu.
     # tat=[],
     # tgk=["tg-TJ"], # Deprecated
-    tgl=["fil-PH"], # "tl-PH" deprecated
-    tha=["th-TH"],
+    tgl=["fil-PH"], # "tl-PH" deprecated, Tagalog
+    fil=["fil-PH"], # Filipino (Standardized form of Tagalog)
+    tha=["th-TH"], # Thai
     # tir=[],
     # tpi=["tpi-PG"], # Deprecated
-    tur=["tr-TR"],
-    ukr=["uk-UA"],
-    urd=["ur-IN"],
-    uzb=["uz-UZ"],
-    vie=["vi-VN"],
-    cmn=["zh-CN", "zh-CN-shandong", "zh-CN-sichuan", "zh-HK", "zh-TW"],
-    zho=["zh-CN", "zh-CN-shandong", "zh-CN-sichuan", "zh-HK", "zh-TW"],
+    tur=["tr-TR"], # Turkish
+    ukr=["uk-UA"], # Ukrainian
+    urd=["ur-IN"], # Urdu
+    uzb=["uz-UZ"], # Uzbek
+    vie=["vi-VN"], # Vietnamese
+    cmn=["zh-CN", "zh-CN-shandong", "zh-CN-sichuan", "zh-HK", "zh-TW"], # Mandarin Chinese
+    zho=["zh-CN", "zh-CN-shandong", "zh-CN-sichuan", "zh-HK", "zh-TW"], # Chinese
     yue=["yue-CN", "zh-HK"], # Cantonese
-    wuu=["wuu-CN"],
+    wuu=["wuu-CN"], # Chinese (Wu, Simplified)
     nan=["zh-TW"], # nan-TW deprecated
     # Note, Taiwanese has one standard + one major dialect,
     # not sure which is covered better by Azure.
-    zul=["zu-ZA"]
+    zul=["zu-ZA"] # Zulu
 )