Bump version 0.5.1

Set compute_type based on device capability.

README.md

@@ -36,7 +36,7 @@ python3 run_server.py --port 9090 \
   
 # running with custom model
 python3 run_server.py --port 9090 \
-                      --backend faster_whisper
+                      --backend faster_whisper \
                       -fw "/path/to/custom/faster/whisper/model"
 ```
 

requirements/server.txt

@@ -1,5 +1,5 @@
 faster-whisper==1.0.1
-torch
+torch==2.3.0
 websockets
 onnxruntime==1.16.0
 numba
@@ -9,4 +9,5 @@ soundfile
 ffmpeg-python
 scipy
 jiwer
-evaluate
+evaluate
+numpy<2

whisper_live/__version__.py

@@ -1 +1 @@
-__version__ = "0.5.0"
+__version__ = "0.5.1"

whisper_live/client.py

@@ -2,6 +2,7 @@ import os
 import shutil
 import wave
 
+import logging
 import numpy as np
 import pyaudio
 import threading
@@ -28,7 +29,8 @@ class Client:
         translate=False,
         model="small",
         srt_file_path="output.srt",
-        use_vad=True
+        use_vad=True,
+        log_transcription=True
     ):
         """
         Initializes a Client instance for audio recording and streaming to a server.
@@ -56,11 +58,11 @@ class Client:
         self.use_vad = use_vad
         self.last_segment = None
         self.last_received_segment = None
+        self.log_transcription = log_transcription
 
         if translate:
             self.task = "translate"
 
-        self.timestamp_offset = 0.0
         self.audio_bytes = None
 
         if host is not None and port is not None:
@@ -117,10 +119,11 @@ class Client:
             self.last_response_received = time.time()
             self.last_received_segment = segments[-1]["text"]
 
-        # Truncate to last 3 entries for brevity.
-        text = text[-3:]
-        utils.clear_screen()
-        utils.print_transcript(text)
+        if self.log_transcription:
+            # Truncate to last 3 entries for brevity.
+            text = text[-3:]
+            utils.clear_screen()
+            utils.print_transcript(text)
 
     def on_message(self, ws, message):
         """
@@ -431,6 +434,8 @@ class TranscriptionTeeClient:
 
     def handle_ffmpeg_process(self, process, stream_type):
         print(f"[INFO]: Connecting to {stream_type} stream...")
+        stderr_thread = threading.Thread(target=self.consume_stderr, args=(process,))
+        stderr_thread.start()
         try:
             # Process the stream
             while True:
@@ -477,6 +482,16 @@ class TranscriptionTeeClient:
 
         return process
 
+    def consume_stderr(self, process):
+        """
+        Consume and log the stderr output of a process in a separate thread.
+
+        Args:
+            process (subprocess.Popen): The process whose stderr output will be logged.
+        """
+        for line in iter(process.stderr.readline, b""):
+            logging.debug(f'[STDERR]: {line.decode()}')
+
     def save_chunk(self, n_audio_file):
         """
         Saves the current audio frames to a WAV file in a separate thread.
@@ -664,9 +679,10 @@ class TranscriptionClient(TranscriptionTeeClient):
         use_vad=True,
         save_output_recording=False,
         output_recording_filename="./output_recording.wav",
-        output_transcription_path="./output.srt"
+        output_transcription_path="./output.srt",
+        log_transcription=True,
     ):
-        self.client = Client(host, port, lang, translate, model, srt_file_path=output_transcription_path, use_vad=use_vad)
+        self.client = Client(host, port, lang, translate, model, srt_file_path=output_transcription_path, use_vad=use_vad, log_transcription=log_transcription)
         if save_output_recording and not output_recording_filename.endswith(".wav"):
             raise ValueError(f"Please provide a valid `output_recording_filename`: {output_recording_filename}")
         if not output_transcription_path.endswith(".srt"):

whisper_live/server.py

@@ -4,6 +4,9 @@ import threading
 import json
 import functools
 import logging
+from enum import Enum
+from typing import List, Optional
+
 import torch
 import numpy as np
 from websockets.sync.server import serve
@@ -121,6 +124,25 @@ class ClientManager:
         return False
 
 
+class BackendType(Enum):
+    FASTER_WHISPER = "faster_whisper"
+    TENSORRT = "tensorrt"
+
+    @staticmethod
+    def valid_types() -> List[str]:
+        return [backend_type.value for backend_type in BackendType]
+
+    @staticmethod
+    def is_valid(backend: str) -> bool:
+        return backend in BackendType.valid_types()
+
+    def is_faster_whisper(self) -> bool:
+        return self == BackendType.FASTER_WHISPER
+
+    def is_tensorrt(self) -> bool:
+        return self == BackendType.TENSORRT
+
+
 class TranscriptionServer:
     RATE = 16000
 
@@ -134,7 +156,9 @@ class TranscriptionServer:
         self, websocket, options, faster_whisper_custom_model_path,
         whisper_tensorrt_path, trt_multilingual
     ):
-        if self.backend == "tensorrt":
+        client: Optional[ServeClientBase] = None
+
+        if self.backend.is_tensorrt():
             try:
                 client = ServeClientTensorRT(
                     websocket,
@@ -155,9 +179,9 @@ class TranscriptionServer:
                     "message": "TensorRT-LLM not supported on Server yet. "
                                "Reverting to available backend: 'faster_whisper'"
                 }))
-                self.backend = "faster_whisper"
+                self.backend = BackendType.FASTER_WHISPER
 
-        if self.backend == "faster_whisper":
+        if self.backend.is_faster_whisper():
             if faster_whisper_custom_model_path is not None and os.path.exists(faster_whisper_custom_model_path):
                 logging.info(f"Using custom model {faster_whisper_custom_model_path}")
                 options["model"] = faster_whisper_custom_model_path
@@ -174,6 +198,9 @@ class TranscriptionServer:
             )
             logging.info("Running faster_whisper backend.")
 
+        if client is None:
+            raise ValueError(f"Backend type {self.backend.value} not recognised or not handled.")
+
         self.client_manager.add_client(websocket, client)
 
     def get_audio_from_websocket(self, websocket):
@@ -202,7 +229,7 @@ class TranscriptionServer:
                 websocket.close()
                 return False  # Indicates that the connection should not continue
 
-            if self.backend == "tensorrt":
+            if self.backend.is_tensorrt():
                 self.vad_detector = VoiceActivityDetector(frame_rate=self.RATE)
             self.initialize_client(websocket, options, faster_whisper_custom_model_path,
                                    whisper_tensorrt_path, trt_multilingual)
@@ -221,11 +248,11 @@ class TranscriptionServer:
         frame_np = self.get_audio_from_websocket(websocket)
         client = self.client_manager.get_client(websocket)
         if frame_np is False:
-            if self.backend == "tensorrt":
+            if self.backend.is_tensorrt():
                 client.set_eos(True)
             return False
 
-        if self.backend == "tensorrt":
+        if self.backend.is_tensorrt():
             voice_active = self.voice_activity(websocket, frame_np)
             if voice_active:
                 self.no_voice_activity_chunks = 0
@@ -238,7 +265,7 @@ class TranscriptionServer:
 
     def recv_audio(self,
                    websocket,
-                   backend="faster_whisper",
+                   backend: BackendType = BackendType.FASTER_WHISPER,
                    faster_whisper_custom_model_path=None,
                    whisper_tensorrt_path=None,
                    trt_multilingual=False):
@@ -311,10 +338,12 @@ class TranscriptionServer:
                 # TODO: load model initially
             else:
                 logging.info("Single model mode currently only works with custom models.")
+        if not BackendType.is_valid(backend):
+            raise ValueError(f"{backend} is not a valid backend type. Choose backend from {BackendType.valid_types()}")
         with serve(
             functools.partial(
                 self.recv_audio,
-                backend=backend,
+                backend=BackendType(backend),
                 faster_whisper_custom_model_path=faster_whisper_custom_model_path,
                 whisper_tensorrt_path=whisper_tensorrt_path,
                 trt_multilingual=trt_multilingual
@@ -758,9 +787,15 @@ class ServeClientFasterWhisper(ServeClientBase):
         self.no_speech_thresh = 0.45
 
         device = "cuda" if torch.cuda.is_available() else "cpu"
+        if device == "cuda":
+            major, _ = torch.cuda.get_device_capability(device)
+            self.compute_type = "float16" if major >= 7 else "float32"
+        else:
+            self.compute_type = "int8"
 
         if self.model_size_or_path is None:
             return
+        logging.info(f"Using Device={device} with precision {self.compute_type}")
 
         if single_model:
             if ServeClientFasterWhisper.SINGLE_MODEL is None:
@@ -793,7 +828,7 @@ class ServeClientFasterWhisper(ServeClientBase):
         self.transcriber = WhisperModel(
             self.model_size_or_path,
             device=device,
-            compute_type="int8" if device == "cpu" else "float16",
+            compute_type=self.compute_type,
             local_files_only=False,
         )
 
@@ -944,6 +979,7 @@ class ServeClientFasterWhisper(ServeClientBase):
 
             input_bytes, duration = self.get_audio_chunk_for_processing()
             if duration < 1.0:
+                time.sleep(0.1)     # wait for audio chunks to arrive
                 continue
             try:
                 input_sample = input_bytes.copy()
@@ -1002,6 +1038,8 @@ class ServeClientFasterWhisper(ServeClientBase):
         """
         offset = None
         self.current_out = ''
+        last_segment = None
+
         # process complete segments
         if len(segments) > 1:
             for i, s in enumerate(segments[:-1]):
@@ -1017,12 +1055,14 @@ class ServeClientFasterWhisper(ServeClientBase):
                 self.transcript.append(self.format_segment(start, end, text_))
                 offset = min(duration, s.end)
 
-        self.current_out += segments[-1].text
-        last_segment = self.format_segment(
-            self.timestamp_offset + segments[-1].start,
-            self.timestamp_offset + min(duration, segments[-1].end),
-            self.current_out
-        )
+        # only process the segments if it satisfies the no_speech_thresh
+        if segments[-1].no_speech_prob <= self.no_speech_thresh:
+            self.current_out += segments[-1].text
+            last_segment = self.format_segment(
+                self.timestamp_offset + segments[-1].start,
+                self.timestamp_offset + min(duration, segments[-1].end),
+                self.current_out
+            )
 
         # if same incomplete segment is seen multiple times then update the offset
         # and append the segment to the list

whisper_live/vad.py

@@ -94,7 +94,7 @@ class VoiceActivityDetection():
         return stacked.cpu()
 
     @staticmethod
-    def download(model_url="https://github.com/snakers4/silero-vad/raw/master/files/silero_vad.onnx"):
+    def download(model_url="https://github.com/snakers4/silero-vad/raw/v4.0/files/silero_vad.onnx"):
         target_dir = os.path.expanduser("~/.cache/whisper-live/")
 
         # Ensure the target directory exists