types::untested vlm path rn

llmpeg/capabilities/audio/audio.py

@@ -1,12 +1,14 @@
 from pathlib import Path
 from dataclasses import dataclass
 from typing import Union
+from io import BytesIO
 
 import numpy as np
 
 from llmpeg.capabilities.audio.audio_input import AudioInput
 from llmpeg.capabilities.audio.audio_output import AudioOutput
-
+from llmpeg.capabilities.filesystem import WaveFile
+from llmpeg.types import CurrentDate, URL, Error
 
 @dataclass
 class Audio:
@@ -21,20 +23,39 @@ def __post_init__(self):
     def capture_stream(self, duration: int):
         return self.audio_input.capture_stream(duration)
 
-    def capture_to_file(self, duration: int):
-        return self.audio_input.capture_to_file(duration)
+    def capture_to_file(self, duration: int = 5, sr: int = 16000) -> Path:
+        audio_stream = self.capture_stream(duration, sr)
+        audio_file = self.cache_dir / f'{CurrentDate().date}.wav'
+        WaveFile.write(audio_stream, audio_file, sr)
+        return audio_file
 
     def play_audio_stream(self, audio_stream: Union[bytes, np.float32]) -> None:
-        self.audio_output.play([audio_stream])
+        self.audio_output.play([self.process_track(audio_stream)])
 
     def play_remote_audio_stream_url(self, url: str) -> None:
-        self.audio_output.play([url])
+        self.audio_output.play([self.process_track(URL(url))])
 
     def play_audio_file(self, audio_file: Path) -> None:
-        self.audio_output.play([audio_file])
+        self.audio_output.play([self.process_track(audio_file)])
 
     def play(self, tracks: list[Union[str, Path, bytes, np.float32]]) -> None:
-        self.audio_output.play(tracks)
+        self.audio_output.play([self.process_track(track) for track in tracks])
 
     def stop(self):
         self.audio_output.stop()
+
+    def process_track(self, track: Union[URL, Path, bytes, np.float32]) -> bytes:
+        if isinstance(track, (str, Path)):
+            _, data = WaveFile.read(track)
+        elif isinstance(track, bytes):
+            _, data = WaveFile.read(BytesIO(track))
+        elif isinstance(track, np.ndarray):
+            _ = 44100  # Assuming sample rate of 44100 Hz
+            data = (track * np.iinfo(np.int16).max).astype(np.int16)
+        elif isinstance(track, np.float32):
+            _ = 44100
+        else:
+            err = Error('Unsupported audio format').__repr__()
+            raise ValueError(err)
+        return data.tobytes()
+

llmpeg/capabilities/audio/audio_input.py

@@ -4,7 +4,8 @@
 from pathlib import Path
 from functools import partial
 
-from llmpeg.utils import CurrentDate, WaveFile
+from llmpeg.types import CurrentDate
+from llmpeg.capabilities.filesystem import WaveFile
 
 
 @dataclass
@@ -14,7 +15,7 @@ class AudioInput:
     def __post_init__(self):
         self.audio = pyaudio.PyAudio()
 
-    def capture_stream(self, duration: int = 5, sr: int = 16000) -> np.float32:
+    def capture_stream(self, duration: int, sr: int) -> np.float32:
         CHUNK = 1024
         FORMAT = pyaudio.paInt16  # int16
         CHANNELS = 1
@@ -41,8 +42,3 @@ def capture_stream(self, duration: int = 5, sr: int = 16000) -> np.float32:
     def __del__(self):
         self.audio.terminate()
 
-    def capture_to_file(self, duration: int = 5, sr: int = 16000) -> Path:
-        audio_stream = self.capture_stream(duration, sr)
-        audio_file = self.cache_dir / f'{CurrentDate().date}.wav'
-        WaveFile.write(audio_stream, audio_file, sr)
-        return audio_file

llmpeg/capabilities/audio/audio_output.py

@@ -3,12 +3,12 @@
 import queue
 from typing import Union
 from pathlib import Path
-from io import BytesIO
 
 import numpy as np
 import pyaudio
 
-from llmpeg.utils import WaveFile, Error
+from llmpeg.types import Error
+from llmpeg.capabilities.filesystem import WaveFile
 
 
 class AudioOutput:
@@ -21,28 +21,17 @@ def __post_init__(self) -> None:
         self.queue = queue.Queue()
         self.pa = pyaudio.PyAudio()
 
-    def _play_audio(self, track: Union[str, Path, bytes, np.float32]) -> None:
+
+    def _play_audio(self, track: Union[bytes, np.float32]) -> None:
         # 16-bit signed integer format
         stream = self.pa.open(
             format=pyaudio.paInt16,
             channels=1,
             rate=44100,
             output=True,
         )
-
-        if isinstance(track, (str, Path)):
-            _, data = WaveFile.read(track)
-        elif isinstance(track, bytes):
-            _, data = WaveFile.read(BytesIO(track))
-        elif isinstance(track, np.ndarray):
-            _ = 44100  # Assuming sample rate of 44100 Hz
-            data = (track * np.iinfo(np.int16).max).astype(np.int16)
-        else:
-            err = Error('Unsupported audio format').__repr__()
-            raise ValueError(err)
-
         self.playing = True
-        stream.write(data.tobytes())
+        stream.write(track.tobytes())
         stream.stop_stream()
         stream.close()
         self.playing = False
@@ -51,23 +40,24 @@ def stop(self) -> None:
         self.stop_event.set()
         self.queue.queue.clear()
 
-    def play(self, tracks: list[Union[str, Path, bytes, np.float32]]) -> None:
+    def play(self, tracks: list[Union[bytes, np.float32]]) -> None:
         self.stop()
         self.stop_event.clear()
         self.playing = True
-        for track in tracks:
-            if track:
-                self.queue.put(track)
+        map(lambda track: self.queue.put(track), [track for track in tracks if track])
         self.thread = threading.Thread(target=self._play_from_queue)
         self.thread.start()
 
+    def _play(self) -> None:
+        if not self.queue.empty():
+            track = self.queue.get()
+            try:
+                self._play_audio(track)
+            except Exception as e:
+                raise Exception(Error(e).__repr__())
+        else:
+            time.sleep(0.1)
+
     def _play_from_queue(self) -> None:
         while not self.stop_event.is_set():
-            if not self.queue.empty():
-                track = self.queue.get()
-                try:
-                    self._play_audio(track)
-                except Exception as e:
-                    raise Exception(Error(e).__repr__())
-            else:
-                time.sleep(0.1)
+            self._play()

llmpeg/capabilities/filesystem/filesystem.py

@@ -11,8 +11,7 @@ class FileCacheDirectory:
 
     def __post_init__(self):
         if not self.cache_dir:
-            self.cache_dir = Path(f'~/.cache/{str(
-                Path(__file__).cwd().name).split("/")[-1]}').expanduser()
+            self.cache_dir = Path(f'~/.cache/{str(Path(__file__).cwd().name).split("/")[-1]}').expanduser()
         Path.mkdir(self.cache_dir, exist_ok=True)
 
     def __str__(self) -> Path: