support whisper's turbo model for transcription

Pipfile

@@ -55,7 +55,7 @@ mccabe = "==0.6.1"
 numba = ">=0.50.0"
 numpy = "<1.24.0"
 oauthlib = "==3.1.0"
-openai-whisper = "==20231117"
+openai-whisper = "==20240930"
 pbr = "==4.0.2"
 pkgconfig = "~=1.5.5"
 pluggy = "==0.13.1"
@@ -78,6 +78,7 @@ rsa = "==4.7"
 scipy = "<1.11.0"
 scikit-learn = ">=0.19.1"
 sentencepiece = "~=0.1.95"
+setuptools = "<65.0.0"
 six = "~=1.15.0"
 tensorflow = ">=1.15.5,<2.12"
 termcolor = "==1.1.0"

pyproject.toml

@@ -1,5 +1,5 @@
 [build-system]
-requires = ["setuptools", "wheel", "Cython"]
+requires = ["setuptools<65.0.0", "wheel", "Cython"]
 
 [tool.pydoclint]
 style = 'sphinx'

requirements-dev.txt

@@ -10,7 +10,7 @@ radish-bdd~=0.13.3
 pex<=2.1.80
 mypy==1.3.0
 types-requests==2.27.9
-types-setuptools==57.4.9
+types-setuptools==64.0.1
 typing-extensions==4.5.0
 parameterized==0.8.1
 pylint~=2.17.2

requirements-llm.txt

@@ -1,4 +1,4 @@
 sentencepiece~=0.1.95
 torch<2.3.0
 transformers<4.37.0
-openai-whisper==20231117
+openai-whisper==20240930

requirements.txt

@@ -48,6 +48,7 @@ PyYAML>=4.2b1
 rsa==4.7
 scipy<1.12.0
 scikit-learn<1.2.0
+setuptools<65.0.0
 six~=1.15.0
 tensorflow>=1.15.5,<2.13
 termcolor==1.1.0

subaligner/__main__.py

@@ -4,7 +4,7 @@
                   [-sil {afr,amh,ara,arg,asm,aze,ben,bos,bul,cat,ces,cmn,cym,dan,deu,ell,eng,epo,est,eus,fas,fin,fra,gla,gle,glg,grc,grn,guj,heb,hin,hrv,hun,hye,ina,ind,isl,ita,jbo,jpn,kal,kan,kat,kir,kor,kur,lat,lav,lfn,lit,mal,mar,mkd,mlt,msa,mya,nah,nep,nld,nor,ori,orm,pan,pap,pol,por,ron,rus,sin,slk,slv,spa,sqi,srp,swa,swe,tam,tat,tel,tha,tsn,tur,ukr,urd,vie,yue,zho}]
                   [-fos] [-tod TRAINING_OUTPUT_DIRECTORY] [-o OUTPUT] [-t TRANSLATE] [-os OFFSET_SECONDS]
                   [-ml {afr,amh,ara,arg,asm,aze,ben,bos,bul,cat,ces,cmn,cym,dan,deu,ell,eng,epo,est,eus,fas,fin,fra,gla,gle,glg,grc,grn,guj,heb,hin,hrv,hun,hye,ina,ind,isl,ita,jbo,jpn,kal,kan,kat,kir,kor,kur,lat,lav,lfn,lit,mal,mar,mkd,mlt,msa,mya,nah,nep,nld,nor,ori,orm,pan,pap,pol,por,ron,rus,sin,slk,slv,spa,sqi,srp,swa,swe,tam,tat,tel,tha,tsn,tur,ukr,urd,vie,yue,zho}]
-                  [-mr {whisper}] [-mf {tiny,tiny.en,small,medium,medium.en,base,base.en,large-v1,large-v2,large-v3,large}] [-tr {helsinki-nlp,whisper,facebook-mbart}] [-tf TRANSLATION_FLAVOUR]
+                  [-mr {whisper}] [-mf {tiny,tiny.en,small,medium,medium.en,base,base.en,large-v1,large-v2,large-v3,large,turbo}] [-tr {helsinki-nlp,whisper,facebook-mbart}] [-tf TRANSLATION_FLAVOUR]
                   [-mpt MEDIA_PROCESS_TIMEOUT] [-sat SEGMENT_ALIGNMENT_TIMEOUT] [-lgs] [-d] [-q] [-ver]
 
 Subaligner command line interface (v0.3.7)
@@ -32,7 +32,7 @@
                         Target video's main language as an ISO 639-3 language code [https://en.wikipedia.org/wiki/List_of_ISO_639-3_codes]
   -mr {whisper}, --transcription_recipe {whisper}
                         LLM recipe used for transcribing video files
-  -mf {tiny,tiny.en,small,medium,medium.en,base,base.en,large-v1,large-v2,large-v3,large}, --transcription_flavour {tiny,tiny.en,small,medium,medium.en,base,base.en,large-v1,large-v2,large-v3,large}
+  -mf {tiny,tiny.en,small,medium,medium.en,base,base.en,large-v1,large-v2,large-v3,large,turbo}, --transcription_flavour {tiny,tiny.en,small,medium,medium.en,base,base.en,large-v1,large-v2,large-v3,large}
                         Flavour variation for a specific LLM recipe supporting transcription
   -tr {helsinki-nlp,whisper,facebook-mbart}, --translation_recipe {helsinki-nlp,whisper,facebook-mbart}
                         LLM recipe used for translating subtitles

subaligner/llm.py

@@ -23,6 +23,7 @@ class WhisperFlavour(Enum):
     LARGE_V2 = "large-v2"
     LARGE_V3 = "large-v3"
     LARGE = "large"
+    TURBO = "turbo"
 
 
 class HelsinkiNLPFlavour(Enum):

subaligner/subaligner_batch/__main__.py

@@ -4,7 +4,7 @@
                         [-sil {afr,amh,ara,arg,asm,aze,ben,bos,bul,cat,ces,cmn,cym,dan,deu,ell,eng,epo,est,eus,fas,fin,fra,gla,gle,glg,grc,grn,guj,heb,hin,hrv,hun,hye,ina,ind,isl,ita,jbo,jpn,kal,kan,kat,kir,kor,kur,lat,lav,lfn,lit,mal,mar,mkd,mlt,msa,mya,nah,nep,nld,nor,ori,orm,pan,pap,pol,por,ron,rus,sin,slk,slv,spa,sqi,srp,swa,swe,tam,tat,tel,tha,tsn,tur,ukr,urd,vie,yue,zho}]
                         [-fos] [-tod TRAINING_OUTPUT_DIRECTORY] [-od OUTPUT_DIRECTORY] [-of {srt,ytt,ttml,txt,smi,xml,ssa,ass,dfxp,sub,scc,tmp,sami,vtt,stl,sbv}] [-t TRANSLATE]
                         [-ml {afr,amh,ara,arg,asm,aze,ben,bos,bul,cat,ces,cmn,cym,dan,deu,ell,eng,epo,est,eus,fas,fin,fra,gla,gle,glg,grc,grn,guj,heb,hin,hrv,hun,hye,ina,ind,isl,ita,jbo,jpn,kal,kan,kat,kir,kor,kur,lat,lav,lfn,lit,mal,mar,mkd,mlt,msa,mya,nah,nep,nld,nor,ori,orm,pan,pap,pol,por,ron,rus,sin,slk,slv,spa,sqi,srp,swa,swe,tam,tat,tel,tha,tsn,tur,ukr,urd,vie,yue,zho}]
-                        [-mr {whisper}] [-mf {tiny,tiny.en,small,medium,medium.en,base,base.en,large-v1,large-v2,large-v3,large}] [-lgs] [-d] [-q] [-ver]
+                        [-mr {whisper}] [-mf {tiny,tiny.en,small,medium,medium.en,base,base.en,large-v1,large-v2,large-v3,large,turbo}] [-lgs] [-d] [-q] [-ver]
 
 Batch align multiple subtitle files and audiovisual files
 
@@ -36,7 +36,7 @@
                         Target video's main language as an ISO 639-3 language code [https://en.wikipedia.org/wiki/List_of_ISO_639-3_codes]
   -mr {whisper}, --transcription_recipe {whisper}
                         LLM recipe used for transcribing video files
-  -mf {tiny,tiny.en,small,medium,medium.en,base,base.en,large-v1,large-v2,large-v3,large}, --transcription_flavour {tiny,tiny.en,small,medium,medium.en,base,base.en,large-v1,large-v2,large-v3,large}
+  -mf {tiny,tiny.en,small,medium,medium.en,base,base.en,large-v1,large-v2,large-v3,large,turbo}, --transcription_flavour {tiny,tiny.en,small,medium,medium.en,base,base.en,large-v1,large-v2,large-v3,large}
                         Flavour variation for a specific LLM recipe supporting transcription
   -lgs, --languages     Print out language codes used for stretch and translation
   -d, --debug           Print out debugging information