Toak/Core/TranscriptionOrchestrator.cs

using System;
using System.Diagnostics;
using System.IO;
using System.Net.Sockets;
using System.Threading.Tasks;
using Toak.Core.Interfaces;
using Toak.Configuration;

namespace Toak.Core;

public class TranscriptionOrchestrator : ITranscriptionOrchestrator
{
    private readonly ISpeechClient _speechClient;
    private readonly ILlmClient _llmClient;
    private readonly IConfigProvider _configProvider;
    private readonly IAudioRecorder _audioRecorder;
    private readonly INotifications _notifications;
    private readonly ITextInjector _textInjector;
    private readonly IHistoryManager _historyManager;
    private readonly IClipboardManager _clipboardManager;
    private readonly IRecordingStateTracker _stateTracker;

    public TranscriptionOrchestrator(
        ISpeechClient speechClient,
        ILlmClient llmClient,
        IConfigProvider configProvider,
        IAudioRecorder audioRecorder,
        INotifications notifications,
        ITextInjector textInjector,
        IHistoryManager historyManager,
        IClipboardManager clipboardManager,
        IRecordingStateTracker stateTracker)
    {
        _speechClient = speechClient;
        _llmClient = llmClient;
        _configProvider = configProvider;
        _audioRecorder = audioRecorder;
        _notifications = notifications;
        _textInjector = textInjector;
        _historyManager = historyManager;
        _clipboardManager = clipboardManager;
        _stateTracker = stateTracker;
    }

    public async Task ProcessStartRecordingAsync()
    {
        if (_stateTracker.IsRecording()) return;

        Logger.LogDebug("Received START command");
        var config = _configProvider.LoadConfig();
        _notifications.PlaySound(config.StartSoundPath);
        _audioRecorder.StartRecording();
    }

    public async Task ProcessStopRecordingAsync(Socket client, bool pipeToStdout, bool copyToClipboard)
    {
        if (!_stateTracker.IsRecording()) return;

        Logger.LogDebug("Received STOP command");
        var config = _configProvider.LoadConfig();
        _notifications.PlaySound(config.StopSoundPath);
        _notifications.Notify("Toak", "Transcribing...");
        
        _audioRecorder.StopRecording();

        var wavPath = _audioRecorder.GetWavPath();
        if (!File.Exists(wavPath) || new FileInfo(wavPath).Length == 0)
        {
            _notifications.Notify("Toak", "No audio recorded.");
            return;
        }

        try
        {
            var stopWatch = Stopwatch.StartNew();
            
            Logger.LogDebug($"Starting STT via Whisper for {wavPath}...");
            var transcript = await _speechClient.TranscribeAsync(wavPath, config.WhisperLanguage, config.WhisperModel);
            
            if (string.IsNullOrWhiteSpace(transcript))
            {
                _notifications.Notify("Toak", "No speech detected.");
                return;
            }

            var detectedSkill = Toak.Core.Skills.SkillRegistry.DetectSkill(transcript, config.ActiveSkills);
            string systemPrompt = detectedSkill != null ? detectedSkill.GetSystemPrompt(transcript) : PromptBuilder.BuildPrompt(config);
            bool isExecutionSkill = detectedSkill != null && detectedSkill.HandlesExecution;

            if (isExecutionSkill)
            {
                var finalText = await _llmClient.RefineTextAsync(transcript, systemPrompt, config.LlmModel);
                if (!string.IsNullOrWhiteSpace(finalText))
                {
                    detectedSkill!.Execute(finalText);
                    stopWatch.Stop();
                    _historyManager.SaveEntry(transcript, finalText, detectedSkill.Name, stopWatch.ElapsedMilliseconds);
                    _notifications.Notify("Toak", $"Skill executed in {stopWatch.ElapsedMilliseconds}ms");
                }
            }
            else
            {
                Logger.LogDebug("Starting LLM text refinement (streaming)...");
                var tokenStream = _llmClient.RefineTextStreamAsync(transcript, systemPrompt, config.LlmModel);
                
                if (pipeToStdout || copyToClipboard)
                {
                    string fullText = "";
                    await foreach (var token in tokenStream)
                    {
                        fullText += token;
                        if (pipeToStdout)
                        {
                            await client.SendAsync(System.Text.Encoding.UTF8.GetBytes(token), SocketFlags.None);
                        }
                    }
                    stopWatch.Stop();
                    if (copyToClipboard)
                    {
                        _clipboardManager.Copy(fullText);
                        _notifications.Notify("Toak", $"Copied to clipboard in {stopWatch.ElapsedMilliseconds}ms");
                    }
                    _historyManager.SaveEntry(transcript, fullText, detectedSkill?.Name, stopWatch.ElapsedMilliseconds);
                }
                else
                {
                    string fullText = await _textInjector.InjectStreamAsync(tokenStream, config.TypingBackend);
                    stopWatch.Stop();
                    _historyManager.SaveEntry(transcript, fullText, detectedSkill?.Name, stopWatch.ElapsedMilliseconds);
                    _notifications.Notify("Toak", $"Done in {stopWatch.ElapsedMilliseconds}ms");
                }
            }
        }
        catch (Exception ex)
        {
            _notifications.Notify("Toak Error", ex.Message);
            Logger.LogDebug($"Error during processing: {ex.Message}");
        }
        finally
        {
            if (File.Exists(wavPath)) File.Delete(wavPath);
        }
    }

    public void ProcessAbortAsync()
    {
        Logger.LogDebug("Received ABORT command");
        _audioRecorder.StopRecording();
        var wavPath = _audioRecorder.GetWavPath();
        if (File.Exists(wavPath)) File.Delete(wavPath);
        _notifications.Notify("Toak", "Recording Aborted.");
    }
}