Update modules/vad/silero_vad.py

modules/vad/silero_vad.py

@@ -9,24 +9,26 @@ import faster_whisper
 from faster_whisper.transcribe import SpeechTimestampsMap
 import gradio as gr
 
-class Segment(self):
-    id: Optional[int] = Field(default=None, description="Incremental id for the segment")
-    seek: Optional[int] = Field(default=None, description="Seek of the segment from chunked audio")
-    text: Optional[str] = Field(default=None, description="Transcription text of the segment")
-    start: Optional[float] = Field(default=None, description="Start time of the segment")
-    end: Optional[float] = Field(default=None, description="End time of the segment")
-    tokens: Optional[List[int]] = Field(default=None, description="List of token IDs")
-    temperature: Optional[float] = Field(default=None, description="Temperature used during the decoding process")
-    avg_logprob: Optional[float] = Field(default=None, description="Average log probability of the tokens")
-    compression_ratio: Optional[float] = Field(default=None, description="Compression ratio of the segment")
-    no_speech_prob: Optional[float] = Field(default=None, description="Probability that it's not speech")
-    words: Optional[List['Word']] = Field(default=None, description="List of words contained in the segment")
+class Segment:
+    def __init__(self):
+        self.id: Optional[int] = Field(default=None, description="Incremental id for the segment")
+        self.seek: Optional[int] = Field(default=None, description="Seek of the segment from chunked audio")
+        self.text: Optional[str] = Field(default=None, description="Transcription text of the segment")
+        self.start: Optional[float] = Field(default=None, description="Start time of the segment")
+        self.end: Optional[float] = Field(default=None, description="End time of the segment")
+        self.tokens: Optional[List[int]] = Field(default=None, description="List of token IDs")
+        self.temperature: Optional[float] = Field(default=None, description="Temperature used during the decoding process")
+        self.avg_logprob: Optional[float] = Field(default=None, description="Average log probability of the tokens")
+        self.compression_ratio: Optional[float] = Field(default=None, description="Compression ratio of the segment")
+        self.no_speech_prob: Optional[float] = Field(default=None, description="Probability that it's not speech")
+        self.words: Optional[List['Word']] = Field(default=None, description="List of words contained in the segment")
 
 class Word(self):
-    start: Optional[float] = Field(default=None, description="Start time of the word")
-    end: Optional[float] = Field(default=None, description="Start time of the word")
-    word: Optional[str] = Field(default=None, description="Word text")
-    probability: Optional[float] = Field(default=None, description="Probability of the word")
+    def __init__(self):
+        self.start: Optional[float] = Field(default=None, description="Start time of the word")
+        self.end: Optional[float] = Field(default=None, description="Start time of the word")
+        self.word: Optional[str] = Field(default=None, description="Word text")
+        self.probability: Optional[float] = Field(default=None, description="Probability of the word")
 
 class SileroVAD:
     def __init__(self):