Commit @5428d1c1719045c9fd241677117c76cae9d7f0b6 - yjyoon/whisper_streaming

Tijs Zwinkels 2024-02-15

Make OpenAI backend work with language autodetect

@5428d1c1719045c9fd241677117c76cae9d7f0b6

5da3267

5428d1c

whisper_online.py

--- whisper_online.py

+++ whisper_online.py


 class OpenaiApiASR(ASRBase):
     """Uses OpenAI's Whisper API for audio transcription."""
 
-    def __init__(self, lan=None, response_format="verbose_json", temperature=0, logfile=sys.stderr):
+    def __init__(self, lan=None, temperature=0, logfile=sys.stderr):
         self.logfile = logfile
 
         self.modelname = "whisper-1"  
-        self.language = lan  # ISO-639-1 language code
-        self.response_format = response_format
+        self.original_language = None if lan == "auto" else lan # ISO-639-1 language code
+        self.response_format = "verbose_json" 
         self.temperature = temperature
 
         self.load_model()

             "temperature": self.temperature,
             "timestamp_granularities": ["word", "segment"]
         }
-        if self.task != "translate" and self.language:
-            params["language"] = self.language
+        if self.task != "translate" and self.original_language:
+            params["language"] = self.original_language
         if prompt:
             params["prompt"] = prompt
 

Add a comment

Open 0
Closed 0

List

...	...	@@ -152,12 +152,12 @@
152	152	class OpenaiApiASR(ASRBase):
153	153	"""Uses OpenAI's Whisper API for audio transcription."""
154	154
155		- def __init__(self, lan=None, response_format="verbose_json", temperature=0, logfile=sys.stderr):
	155	+ def __init__(self, lan=None, temperature=0, logfile=sys.stderr):
156	156	self.logfile = logfile
157	157
158	158	self.modelname = "whisper-1"
159		- self.language = lan # ISO-639-1 language code
160		- self.response_format = response_format
	159	+ self.original_language = None if lan == "auto" else lan # ISO-639-1 language code
	160	+ self.response_format = "verbose_json"
161	161	self.temperature = temperature
162	162
163	163	self.load_model()
...	...	@@ -213,8 +213,8 @@
213	213	"temperature": self.temperature,
214	214	"timestamp_granularities": ["word", "segment"]
215	215	}
216		- if self.task != "translate" and self.language:
217		- params["language"] = self.language
	216	+ if self.task != "translate" and self.original_language:
	217	+ params["language"] = self.original_language
218	218	if prompt:
219	219	params["prompt"] = prompt
220	220

Delete comment