Commit @1c523f2f0def003b4c2b93b9d03a081022f1d674 - yjyoon/whisper_streaming

Dominik Macháček 2024-02-06

Merge branch 'main' into vad-streaming

@1c523f2f0def003b4c2b93b9d03a081022f1d674

891cb12

1c523f2

README.md

--- README.md

+++ README.md


   --model_dir MODEL_DIR
                         Dir where Whisper model.bin and other files are saved. This option overrides --model and --model_cache_dir parameter.
   --lan LAN, --language LAN
-                        Language code for transcription, e.g. en,de,cs.
+                        Source language code, e.g. en,de,cs, or 'auto' for language detection.
   --task {transcribe,translate}
                         Transcribe or translate.
   --backend {faster-whisper,whisper_timestamped}

891cb12

1c523f2

whisper_online.py

--- whisper_online.py

+++ whisper_online.py


         self.logfile = logfile
 
         self.transcribe_kargs = {}
-        self.original_language = lan 
+        if lan == "auto":
+            self.original_language = None
+        else:
+            self.original_language = lan
 
         self.model = self.load_model(modelsize, cache_dir, model_dir)
 

 
     def transcribe(self, audio, init_prompt=""):
 
-        # tiempo_inicio = datetime.datetime.now()
         # tested: beam_size=5 is faster and better than 1 (on one 200 second document from En ESIC, min chunk 0.01)
         segments, info = self.model.transcribe(audio, language=self.original_language, initial_prompt=init_prompt, beam_size=5, word_timestamps=True, condition_on_previous_text=True, **self.transcribe_kargs)
-        
-        # print(f'({datetime.datetime.now().strftime("%Y-%m-%d %H:%M:%S.%f")})----------r> whisper transcribe  take { (datetime.datetime.now() -tiempo_inicio)  } ms.')
+        #print(info)  # info contains language detection result
 
         return list(segments)
 

     parser.add_argument('--model', type=str, default='large-v2', choices="tiny.en,tiny,base.en,base,small.en,small,medium.en,medium,large-v1,large-v2,large-v3,large".split(","),help="Name size of the Whisper model to use (default: large-v2). The model is automatically downloaded from the model hub if not present in model cache dir.")
     parser.add_argument('--model_cache_dir', type=str, default=None, help="Overriding the default model cache dir where models downloaded from the hub are saved")
     parser.add_argument('--model_dir', type=str, default=None, help="Dir where Whisper model.bin and other files are saved. This option overrides --model and --model_cache_dir parameter.")
-    parser.add_argument('--lan', '--language', type=str, default='en', help="Language code for transcription, e.g. en,de,cs.")
+    parser.add_argument('--lan', '--language', type=str, default='en', help="Source language code, e.g. en,de,cs, or 'auto' for language detection.")
     parser.add_argument('--task', type=str, default='transcribe', choices=["transcribe","translate"],help="Transcribe or translate.")
     parser.add_argument('--backend', type=str, default="faster-whisper", choices=["faster-whisper", "whisper_timestamped"],help='Load only this backend for Whisper processing.')
     parser.add_argument('--vad', action="store_true", default=False, help='Use VAD = voice activity detection, with the default parameters.')

Add a comment

Open 0
Closed 0

List

...	...	@@ -86,7 +86,7 @@
86	86	--model_dir MODEL_DIR
87	87	Dir where Whisper model.bin and other files are saved. This option overrides --model and --model_cache_dir parameter.
88	88	--lan LAN, --language LAN
89		- Language code for transcription, e.g. en,de,cs.
	89	+ Source language code, e.g. en,de,cs, or 'auto' for language detection.
90	90	--task {transcribe,translate}
91	91	Transcribe or translate.
92	92	--backend {faster-whisper,whisper_timestamped}

...	...	@@ -30,7 +30,10 @@
30	30	self.logfile = logfile
31	31
32	32	self.transcribe_kargs = {}
33		- self.original_language = lan
	33	+ if lan == "auto":
	34	+ self.original_language = None
	35	+ else:
	36	+ self.original_language = lan
34	37
35	38	self.model = self.load_model(modelsize, cache_dir, model_dir)
36	39
...	...	@@ -119,11 +122,9 @@
119	122
120	123	def transcribe(self, audio, init_prompt=""):
121	124
122		- # tiempo_inicio = datetime.datetime.now()
123	125	# tested: beam_size=5 is faster and better than 1 (on one 200 second document from En ESIC, min chunk 0.01)
124	126	segments, info = self.model.transcribe(audio, language=self.original_language, initial_prompt=init_prompt, beam_size=5, word_timestamps=True, condition_on_previous_text=True, **self.transcribe_kargs)
125		-
126		- # print(f'({datetime.datetime.now().strftime("%Y-%m-%d %H:%M:%S.%f")})----------r> whisper transcribe take { (datetime.datetime.now() -tiempo_inicio) } ms.')
	127	+ #print(info) # info contains language detection result
127	128
128	129	return list(segments)
129	130
...	...	@@ -458,7 +459,7 @@
458	459	parser.add_argument('--model', type=str, default='large-v2', choices="tiny.en,tiny,base.en,base,small.en,small,medium.en,medium,large-v1,large-v2,large-v3,large".split(","),help="Name size of the Whisper model to use (default: large-v2). The model is automatically downloaded from the model hub if not present in model cache dir.")
459	460	parser.add_argument('--model_cache_dir', type=str, default=None, help="Overriding the default model cache dir where models downloaded from the hub are saved")
460	461	parser.add_argument('--model_dir', type=str, default=None, help="Dir where Whisper model.bin and other files are saved. This option overrides --model and --model_cache_dir parameter.")
461		- parser.add_argument('--lan', '--language', type=str, default='en', help="Language code for transcription, e.g. en,de,cs.")
	462	+ parser.add_argument('--lan', '--language', type=str, default='en', help="Source language code, e.g. en,de,cs, or 'auto' for language detection.")
462	463	parser.add_argument('--task', type=str, default='transcribe', choices=["transcribe","translate"],help="Transcribe or translate.")
463	464	parser.add_argument('--backend', type=str, default="faster-whisper", choices=["faster-whisper", "whisper_timestamped"],help='Load only this backend for Whisper processing.')
464	465	parser.add_argument('--vad', action="store_true", default=False, help='Use VAD = voice activity detection, with the default parameters.')

Delete comment