add test-clean code merge

talkingq-url/utils/ audio_format.py

@@ -0,0 +1,33 @@
+import io
+import wave
+
+
+DEFAULT_SAMPLE_RATE = 16000
+DEFAULT_CHANNELS = 1
+DEFAULT_SAMPLE_WIDTH = 2
+
+
+def detect_audio_format(audio_data: bytes) -> str:
+    if len(audio_data) >= 12 and audio_data[:4] == b"RIFF" and audio_data[8:12] == b"WAVE":
+        return "wav"
+    if audio_data.startswith(b"ID3"):
+        return "mp3"
+    if len(audio_data) >= 2 and audio_data[0] == 0xFF and (audio_data[1] & 0xE0) == 0xE0:
+        return "mp3"
+    return "pcm_s16le_16k_mono"
+
+
+def wrap_pcm_as_wav(
+    audio_data: bytes,
+    *,
+    sample_rate: int = DEFAULT_SAMPLE_RATE,
+    channels: int = DEFAULT_CHANNELS,
+    sample_width: int = DEFAULT_SAMPLE_WIDTH,
+) -> bytes:
+    wav_buffer = io.BytesIO()
+    with wave.open(wav_buffer, "wb") as wav_file:
+        wav_file.setnchannels(channels)
+        wav_file.setsampwidth(sample_width)
+        wav_file.setframerate(sample_rate)
+        wav_file.writeframes(audio_data)
+    return wav_buffer.getvalue()