feat: add --device flag and devices command for mic selection

Lets the user pick an input device by index or name substring. Adds
`cohere devices` for listing. For devices that don't support 16kHz
natively (e.g. Sipeed MicArray hw at 48kHz), captures at the device's
native rate and resamples to 16kHz via scipy.signal.resample_poly.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

src/cohere_transcribe/cli/cli.py

@@ -12,10 +12,20 @@ app = typer.Typer(help="Cohere live transcription — speaks into your keyboard.
 console = Console()
 
 
+def _parse_device(value: str | None):
+    if value is None:
+        return None
+    try:
+        return int(value)
+    except ValueError:
+        return value
+
+
 @app.command()
 def on(
     language: str = typer.Option("en", "--lang", "-l", help="Language code"),
     pause: float = typer.Option(0.3, "--pause", "-p", help="Seconds of silence before sending text"),
+    device: str = typer.Option(None, "--device", "-d", help="Input device index or name substring (see `cohere devices`)"),
     foreground: bool = typer.Option(False, "--fg", help="Run in foreground (don't daemonize)"),
 ):
     """Start transcribing and typing into your focused window."""
@@ -26,7 +36,7 @@ def on(
     if foreground:
         from ..daemon import run_daemon
         console.print("[green]Starting cohere (foreground)...[/green]")
-        run_daemon(language, pause=pause)
+        run_daemon(language, pause=pause, device=_parse_device(device))
         return
 
     console.print("[green]Starting cohere daemon...[/green]")
@@ -34,6 +44,8 @@ def on(
     cmd = [sys.executable, "-m", "cohere_transcribe.daemon_main", "--lang", language]
     if pause != 0.3:
         cmd += ["--pause", str(pause)]
+    if device is not None:
+        cmd += ["--device", device]
     subprocess.Popen(
         cmd,
         start_new_session=True,
@@ -90,20 +102,23 @@ def transcribe(
     stream: bool = typer.Option(False, "--stream", "-s", help="Live streaming mode (prints to terminal)"),
     language: str = typer.Option("en", "--lang", "-l", help="Language code"),
     pause: float = typer.Option(0.3, "--pause", "-p", help="Seconds of silence before sending text"),
+    device: str = typer.Option(None, "--device", "-d", help="Input device index or name substring (see `cohere devices`)"),
 ):
     """One-shot transcription (file, mic, or stream to terminal)."""
     from ..model import load_model, transcribe_audio
     from ..vad import pause_seconds_to_frames
 
+    dev = _parse_device(device)
+
     if stream:
         from ..stream import stream_transcribe
         processor, model = load_model()
-        stream_transcribe(processor, model, language, silence_frames=pause_seconds_to_frames(pause))
+        stream_transcribe(processor, model, language, silence_frames=pause_seconds_to_frames(pause), device=dev)
     elif mic is not None:
         from ..model import record_audio
         processor, model = load_model()
         try:
-            audio = record_audio(mic)
+            audio = record_audio(mic, device=dev)
             console.print("Transcribing...")
             text = transcribe_audio(processor, model, audio, language)
             console.print(f"\n{text}\n")
@@ -122,5 +137,26 @@ def transcribe(
         raise typer.Exit(1)
 
 
+@app.command()
+def devices():
+    """List available audio input devices."""
+    import sounddevice as sd
+
+    default_in = sd.default.device[0]
+    for idx, dev in enumerate(sd.query_devices()):
+        if dev["max_input_channels"] <= 0:
+            continue
+        marker = "[green]*[/green]" if idx == default_in else " "
+        hostapi = sd.query_hostapis(dev["hostapi"])["name"]
+        console.print(
+            f"{marker} [bold]{idx:>2}[/bold]  {dev['name']}  "
+            f"[dim]({dev['max_input_channels']}ch, {int(dev['default_samplerate'])}Hz, {hostapi})[/dim]"
+        )
+    console.print(
+        "\n[dim]Tip: indices can shift between runs on PipeWire. "
+        "Prefer [bold]-d pipewire[/bold] (uses PipeWire's default source) or pass a name substring like [bold]-d Sipeed[/bold].[/dim]"
+    )
+
+
 def main():
     app()