made small fixes in arguments_classes and TTS folder

TTS/chatTTS_handler.py

@@ -64,8 +64,8 @@ class ChatTTSHandler(BaseHandler):
                 audio_chunk = librosa.resample(gen[0], orig_sr=24000, target_sr=16000)
                 audio_chunk = (audio_chunk * 32768).astype(np.int16)[0]
                 while len(audio_chunk) > self.chunk_size:
-                    yield audio_chunk[: self.chunk_size]  # 返回前 chunk_size 字节的数据
-                    audio_chunk = audio_chunk[self.chunk_size :]  # 移除已返回的数据
+                    yield audio_chunk[: self.chunk_size]  # Return the first chunk_size samples of the audio data
+                    audio_chunk = audio_chunk[self.chunk_size :]  # Remove the samples that have already been returned
                 yield np.pad(audio_chunk, (0, self.chunk_size - len(audio_chunk)))
         else:
             wavs = wavs_gen

arguments_classes/language_model_arguments.py

@@ -6,7 +6,7 @@ class LanguageModelHandlerArguments:
     lm_model_name: str = field(
         default="HuggingFaceTB/SmolLM-360M-Instruct",
         metadata={
-            "help": "The pretrained language model to use. Default is 'microsoft/Phi-3-mini-4k-instruct'."
+            "help": "The pretrained language model to use. Default is 'HuggingFaceTB/SmolLM-360M-Instruct'."
         },
     )
     lm_device: str = field(

arguments_classes/mlx_language_model_arguments.py

@@ -6,7 +6,7 @@ class MLXLanguageModelHandlerArguments:
     mlx_lm_model_name: str = field(
         default="mlx-community/SmolLM-360M-Instruct",
         metadata={
-            "help": "The pretrained language model to use. Default is 'microsoft/Phi-3-mini-4k-instruct'."
+            "help": "The pretrained language model to use. Default is 'mlx-community/SmolLM-360M-Instruct'."
         },
     )
     mlx_lm_device: str = field(

arguments_classes/open_api_language_model_arguments.py

@@ -7,7 +7,7 @@ class OpenApiLanguageModelHandlerArguments:
         # default="HuggingFaceTB/SmolLM-360M-Instruct",
         default="deepseek-chat",
         metadata={
-            "help": "The pretrained language model to use. Default is 'microsoft/Phi-3-mini-4k-instruct'."
+            "help": "The pretrained language model to use. Default is 'deepseek-chat'."
         },
     )
     open_api_user_role: str = field(

arguments_classes/parler_tts_arguments.py

@@ -30,13 +30,13 @@ class ParlerTTSHandlerArguments:
     tts_gen_min_new_tokens: int = field(
         default=64,
         metadata={
-            "help": "Maximum number of new tokens to generate in a single completion. Default is 10, which corresponds to ~0.1 secs"
+            "help": "Maximum number of new tokens to generate in a single completion. Default is 64, which corresponds to ~0.64 secs"
         },
     )
     tts_gen_max_new_tokens: int = field(
         default=512,
         metadata={
-            "help": "Maximum number of new tokens to generate in a single completion. Default is 256, which corresponds to ~6 secs"
+            "help": "Maximum number of new tokens to generate in a single completion. Default is 512, which corresponds to ~12 secs"
         },
     )
     description: str = field(
@@ -51,7 +51,7 @@ class ParlerTTSHandlerArguments:
     play_steps_s: float = field(
         default=1.0,
         metadata={
-            "help": "The time interval in seconds for playing back the generated speech in steps. Default is 0.5 seconds."
+            "help": "The time interval in seconds for playing back the generated speech in steps. Default is 1.0 seconds."
         },
     )
     max_prompt_pad_length: int = field(

arguments_classes/vad_arguments.py

@@ -36,7 +36,7 @@ class VADHandlerArguments:
     speech_pad_ms: int = field(
         default=500,
         metadata={
-            "help": "Amount of padding added to the beginning and end of detected speech segments. Measured in milliseconds. Default is 250 ms."
+            "help": "Amount of padding added to the beginning and end of detected speech segments. Measured in milliseconds. Default is 500 ms."
         },
     )
     audio_enhancement: bool = field(