Add speech-to-text feature using Faster Whisper container

Adds a mic button to the terminal UI that captures speech, transcribes
it via a Faster Whisper sidecar container, and injects the text into
the terminal input. Includes settings panel for model selection
(tiny/small/medium), port config, and container lifecycle management.

- stt-container/: Dockerfile + FastAPI server for Whisper transcription
- Rust backend: STT container management, transcribe_audio IPC command
- Frontend: useSTT hook, SttButton, SttSettings, WAV encoder
- CI: Gitea Actions workflow for multi-arch STT image builds

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

app/src-tauri/Cargo.lock

@@ -2345,6 +2345,16 @@ version = "0.3.17"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "6877bb514081ee2a7ff5ef9de3281f14a4dd4bceac4c09388074a6b5df8a139a"
 
+[[package]]
+name = "mime_guess"
+version = "2.0.5"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "f7c44f8e672c00fe5308fa235f821cb4198414e1c77935c1ab6948d3fd78550e"
+dependencies = [
+ "mime",
+ "unicase",
+]
+
 [[package]]
 name = "miniz_oxide"
 version = "0.8.9"
@@ -3454,6 +3464,7 @@ dependencies = [
  "base64 0.22.1",
  "bytes",
  "futures-core",
+ "futures-util",
  "http",
  "http-body",
  "http-body-util",
@@ -3462,6 +3473,7 @@ dependencies = [
  "hyper-util",
  "js-sys",
  "log",
+ "mime_guess",
  "percent-encoding",
  "pin-project-lite",
  "quinn",
@@ -5053,6 +5065,12 @@ dependencies = [
  "unic-common",
 ]
 
+[[package]]
+name = "unicase"
+version = "2.9.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "dbc4bc3a9f746d862c45cb89d705aa10f187bb96c76001afab07a0d35ce60142"
+
 [[package]]
 name = "unicode-ident"
 version = "1.0.24"

app/src-tauri/Cargo.toml

@@ -29,7 +29,7 @@ log = "0.4"
 fern = { version = "0.7", features = ["date-based"] }
 tar = "0.4"
 include_dir = "0.7"
-reqwest = { version = "0.12", default-features = false, features = ["json", "rustls-tls"] }
+reqwest = { version = "0.12", default-features = false, features = ["json", "rustls-tls", "multipart"] }
 iana-time-zone = "0.1"
 sha2 = "0.10"
 axum = { version = "0.8", features = ["ws"] }

app/src-tauri/src/commands/mod.rs

@@ -5,6 +5,7 @@ pub mod help_commands;
 pub mod mcp_commands;
 pub mod project_commands;
 pub mod settings_commands;
+pub mod stt_commands;
 pub mod terminal_commands;
 pub mod update_commands;
 pub mod web_terminal_commands;

app/src-tauri/src/commands/stt_commands.rs

@@ -0,0 +1,92 @@
+use tauri::{AppHandle, Emitter, State};
+
+use crate::docker::stt;
+use crate::models::app_settings::SttStatus;
+use crate::AppState;
+
+#[tauri::command]
+pub async fn get_stt_status(state: State<'_, AppState>) -> Result<SttStatus, String> {
+    let settings = state.settings_store.get();
+    stt::get_stt_status(&settings.stt).await
+}
+
+#[tauri::command]
+pub async fn start_stt(state: State<'_, AppState>) -> Result<SttStatus, String> {
+    let settings = state.settings_store.get();
+    stt::ensure_stt_running(&settings.stt).await
+}
+
+#[tauri::command]
+pub async fn stop_stt() -> Result<(), String> {
+    stt::stop_stt_container().await
+}
+
+#[tauri::command]
+pub async fn build_stt_image(app_handle: AppHandle) -> Result<(), String> {
+    stt::build_stt_image(move |msg| {
+        let _ = app_handle.emit("stt-build-progress", &msg);
+    })
+    .await
+}
+
+#[tauri::command]
+pub async fn pull_stt_image(app_handle: AppHandle) -> Result<(), String> {
+    stt::pull_stt_image(move |msg| {
+        let _ = app_handle.emit("stt-pull-progress", &msg);
+    })
+    .await
+}
+
+#[tauri::command]
+pub async fn transcribe_audio(
+    audio_data: Vec<u8>,
+    state: State<'_, AppState>,
+) -> Result<String, String> {
+    let settings = state.settings_store.get();
+    if !settings.stt.enabled {
+        return Err("STT is not enabled".to_string());
+    }
+
+    let url = format!("http://127.0.0.1:{}/transcribe", settings.stt.port);
+
+    let file_part = reqwest::multipart::Part::bytes(audio_data)
+        .file_name("recording.wav")
+        .mime_str("audio/wav")
+        .map_err(|e| format!("Failed to create multipart: {}", e))?;
+
+    let mut form = reqwest::multipart::Form::new().part("file", file_part);
+
+    if let Some(ref lang) = settings.stt.language {
+        form = form.text("language", lang.clone());
+    }
+
+    let client = reqwest::Client::new();
+    let response = client
+        .post(&url)
+        .multipart(form)
+        .send()
+        .await
+        .map_err(|e| {
+            if e.is_connect() {
+                "STT container is not running. Start it from Settings.".to_string()
+            } else {
+                format!("Transcription request failed: {}", e)
+            }
+        })?;
+
+    if !response.status().is_success() {
+        let status = response.status();
+        let body = response.text().await.unwrap_or_default();
+        return Err(format!("Transcription failed ({}): {}", status, body));
+    }
+
+    let result: serde_json::Value = response
+        .json()
+        .await
+        .map_err(|e| format!("Failed to parse transcription response: {}", e))?;
+
+    result["text"]
+        .as_str()
+        .map(|s| s.to_string())
+        .ok_or_else(|| "No text in transcription response".to_string())
+}

app/src-tauri/src/docker/mod.rs

@@ -3,7 +3,10 @@ pub mod container;
 pub mod image;
 pub mod exec;
 pub mod network;
+pub mod stt;
 
+#[allow(unused_imports)]
+pub use stt::*;
 #[allow(unused_imports)]
 pub use client::*;
 #[allow(unused_imports)]

app/src-tauri/src/docker/stt.rs

@@ -0,0 +1,266 @@
+use bollard::container::{
+    Config, CreateContainerOptions, ListContainersOptions, RemoveContainerOptions,
+    StartContainerOptions, StopContainerOptions,
+};
+use bollard::image::BuildImageOptions;
+use bollard::models::{HostConfig, Mount, MountTypeEnum, PortBinding};
+use futures_util::StreamExt;
+use std::collections::HashMap;
+use std::io::Write;
+
+use super::client::get_docker;
+use crate::models::app_settings::{SttSettings, SttStatus};
+
+const STT_CONTAINER_NAME: &str = "triple-c-stt";
+const STT_MODEL_VOLUME: &str = "triple-c-stt-model-cache";
+const STT_REGISTRY_IMAGE: &str = "ghcr.io/shadowdao/triple-c-stt:latest";
+const STT_LOCAL_IMAGE: &str = "triple-c-stt:latest";
+const STT_DOCKERFILE: &str = include_str!("../../../../stt-container/Dockerfile");
+const STT_SERVER: &str = include_str!("../../../../stt-container/server.py");
+
+pub async fn get_stt_status(settings: &SttSettings) -> Result<SttStatus, String> {
+    let image_exists = super::image::image_exists(STT_REGISTRY_IMAGE).await.unwrap_or(false)
+        || super::image::image_exists(STT_LOCAL_IMAGE).await.unwrap_or(false);
+
+    let (container_exists, running, model) = match find_stt_container().await? {
+        Some((_, state, env_model)) => (true, state == "running", env_model),
+        None => (false, false, settings.model.clone()),
+    };
+
+    Ok(SttStatus {
+        container_exists,
+        running,
+        port: settings.port,
+        model,
+        image_exists,
+    })
+}
+
+async fn find_stt_container() -> Result<Option<(String, String, String)>, String> {
+    let docker = get_docker()?;
+
+    let filters: HashMap<String, Vec<String>> = HashMap::from([(
+        "name".to_string(),
+        vec![format!("/{}", STT_CONTAINER_NAME)],
+    )]);
+
+    let containers = docker
+        .list_containers(Some(ListContainersOptions {
+            all: true,
+            filters,
+            ..Default::default()
+        }))
+        .await
+        .map_err(|e| format!("Failed to list containers: {}", e))?;
+
+    if let Some(container) = containers.first() {
+        let id = container.id.clone().unwrap_or_default();
+        let state = container.state.clone().unwrap_or_default();
+
+        // Extract WHISPER_MODEL from container env
+        let model = container
+            .labels
+            .as_ref()
+            .and_then(|l| l.get("triple-c.stt.model"))
+            .cloned()
+            .unwrap_or_else(|| "tiny".to_string());
+
+        return Ok(Some((id, state, model)));
+    }
+
+    Ok(None)
+}
+
+async fn create_stt_container(settings: &SttSettings) -> Result<String, String> {
+    let docker = get_docker()?;
+
+    // Try local image first, fall back to registry
+    let image = if super::image::image_exists(STT_LOCAL_IMAGE).await.unwrap_or(false) {
+        STT_LOCAL_IMAGE.to_string()
+    } else if super::image::image_exists(STT_REGISTRY_IMAGE).await.unwrap_or(false) {
+        STT_REGISTRY_IMAGE.to_string()
+    } else {
+        return Err("STT image not found. Please build or pull the image first.".to_string());
+    };
+
+    let port_binding = PortBinding {
+        host_ip: Some("127.0.0.1".to_string()),
+        host_port: Some(settings.port.to_string()),
+    };
+
+    let mut port_bindings = HashMap::new();
+    port_bindings.insert(
+        "9876/tcp".to_string(),
+        Some(vec![port_binding]),
+    );
+
+    let host_config = HostConfig {
+        port_bindings: Some(port_bindings),
+        mounts: Some(vec![Mount {
+            target: Some("/root/.cache/huggingface".to_string()),
+            source: Some(STT_MODEL_VOLUME.to_string()),
+            typ: Some(MountTypeEnum::VOLUME),
+            ..Default::default()
+        }]),
+        ..Default::default()
+    };
+
+    let mut labels = HashMap::new();
+    labels.insert(
+        "triple-c.stt.model".to_string(),
+        settings.model.clone(),
+    );
+    labels.insert(
+        "triple-c.stt.port".to_string(),
+        settings.port.to_string(),
+    );
+
+    let config = Config {
+        image: Some(image),
+        env: Some(vec![format!("WHISPER_MODEL={}", settings.model)]),
+        host_config: Some(host_config),
+        labels: Some(labels),
+        ..Default::default()
+    };
+
+    let options = CreateContainerOptions {
+        name: STT_CONTAINER_NAME,
+        ..Default::default()
+    };
+
+    let response = docker
+        .create_container(Some(options), config)
+        .await
+        .map_err(|e| format!("Failed to create STT container: {}", e))?;
+
+    Ok(response.id)
+}
+
+pub async fn ensure_stt_running(settings: &SttSettings) -> Result<SttStatus, String> {
+    let docker = get_docker()?;
+
+    // Check if container exists and if settings match
+    if let Some((id, state, model)) = find_stt_container().await? {
+        let needs_recreate = model != settings.model;
+
+        if needs_recreate {
+            // Settings changed, recreate
+            if state == "running" {
+                docker
+                    .stop_container(&id, None::<StopContainerOptions>)
+                    .await
+                    .map_err(|e| format!("Failed to stop STT container: {}", e))?;
+            }
+            docker
+                .remove_container(
+                    &id,
+                    Some(RemoveContainerOptions {
+                        force: true,
+                        ..Default::default()
+                    }),
+                )
+                .await
+                .map_err(|e| format!("Failed to remove STT container: {}", e))?;
+        } else if state == "running" {
+            return get_stt_status(settings).await;
+        } else {
+            // Container exists but stopped, start it
+            docker
+                .start_container(&id, None::<StartContainerOptions<String>>)
+                .await
+                .map_err(|e| format!("Failed to start STT container: {}", e))?;
+            return get_stt_status(settings).await;
+        }
+    }
+
+    // Create and start new container
+    let id = create_stt_container(settings).await?;
+    docker
+        .start_container(&id, None::<StartContainerOptions<String>>)
+        .await
+        .map_err(|e| format!("Failed to start STT container: {}", e))?;
+
+    get_stt_status(settings).await
+}
+
+pub async fn stop_stt_container() -> Result<(), String> {
+    let docker = get_docker()?;
+
+    if let Some((id, state, _)) = find_stt_container().await? {
+        if state == "running" {
+            docker
+                .stop_container(&id, None::<StopContainerOptions>)
+                .await
+                .map_err(|e| format!("Failed to stop STT container: {}", e))?;
+        }
+    }
+
+    Ok(())
+}
+
+pub async fn pull_stt_image<F>(on_progress: F) -> Result<(), String>
+where
+    F: Fn(String) + Send + 'static,
+{
+    super::image::pull_image(STT_REGISTRY_IMAGE, on_progress).await
+}
+
+pub async fn build_stt_image<F>(on_progress: F) -> Result<(), String>
+where
+    F: Fn(String) + Send + 'static,
+{
+    let docker = get_docker()?;
+
+    let tar_bytes = create_stt_build_context()
+        .map_err(|e| format!("Failed to create STT build context: {}", e))?;
+
+    let options = BuildImageOptions {
+        t: STT_LOCAL_IMAGE,
+        rm: true,
+        forcerm: true,
+        ..Default::default()
+    };
+
+    let mut stream = docker.build_image(options, None, Some(tar_bytes.into()));
+
+    while let Some(result) = stream.next().await {
+        match result {
+            Ok(output) => {
+                if let Some(stream) = output.stream {
+                    on_progress(stream);
+                }
+                if let Some(error) = output.error {
+                    return Err(format!("Build error: {}", error));
+                }
+            }
+            Err(e) => return Err(format!("Build stream error: {}", e)),
+        }
+    }
+
+    Ok(())
+}
+
+fn create_stt_build_context() -> Result<Vec<u8>, std::io::Error> {
+    let mut buf = Vec::new();
+    {
+        let mut archive = tar::Builder::new(&mut buf);
+
+        let mut dockerfile_header = tar::Header::new_gnu();
+        dockerfile_header.set_size(STT_DOCKERFILE.len() as u64);
+        dockerfile_header.set_mode(0o644);
+        dockerfile_header.set_cksum();
+        archive.append_data(&mut dockerfile_header, "Dockerfile", STT_DOCKERFILE.as_bytes())?;
+
+        let mut server_header = tar::Header::new_gnu();
+        server_header.set_size(STT_SERVER.len() as u64);
+        server_header.set_mode(0o644);
+        server_header.set_cksum();
+        archive.append_data(&mut server_header, "server.py", STT_SERVER.as_bytes())?;
+
+        archive.finish()?;
+    }
+
+    let _ = buf.flush();
+    Ok(buf)
+}
+

app/src-tauri/src/lib.rs

@@ -122,6 +122,8 @@ pub fn run() {
                     if let Some(server) = server_guard.take() {
                         server.stop();
                     }
+                    // Stop STT container
+                    let _ = docker::stt::stop_stt_container().await;
                     // Close all exec sessions
                     state.exec_manager.close_all_sessions().await;
                 });
@@ -181,6 +183,13 @@ pub fn run() {
             commands::web_terminal_commands::stop_web_terminal,
             commands::web_terminal_commands::get_web_terminal_status,
             commands::web_terminal_commands::regenerate_web_terminal_token,
+            // STT
+            commands::stt_commands::get_stt_status,
+            commands::stt_commands::start_stt,
+            commands::stt_commands::stop_stt,
+            commands::stt_commands::build_stt_image,
+            commands::stt_commands::pull_stt_image,
+            commands::stt_commands::transcribe_audio,
         ])
         .run(tauri::generate_context!())
         .expect("error while running tauri application");

app/src-tauri/src/models/app_settings.rs

@@ -76,6 +76,48 @@ pub struct AppSettings {
     pub dismissed_image_digest: Option<String>,
     #[serde(default)]
     pub web_terminal: WebTerminalSettings,
+    #[serde(default)]
+    pub stt: SttSettings,
+}
+
+fn default_stt_model() -> String {
+    "tiny".to_string()
+}
+
+fn default_stt_port() -> u16 {
+    9876
+}
+
+#[derive(Debug, Clone, Serialize, Deserialize)]
+pub struct SttSettings {
+    #[serde(default)]
+    pub enabled: bool,
+    #[serde(default = "default_stt_model")]
+    pub model: String,
+    #[serde(default = "default_stt_port")]
+    pub port: u16,
+    #[serde(default)]
+    pub language: Option<String>,
+}
+
+impl Default for SttSettings {
+    fn default() -> Self {
+        Self {
+            enabled: false,
+            model: default_stt_model(),
+            port: 9876,
+            language: None,
+        }
+    }
+}
+
+#[derive(Debug, Clone, Serialize, Deserialize)]
+pub struct SttStatus {
+    pub container_exists: bool,
+    pub running: bool,
+    pub port: u16,
+    pub model: String,
+    pub image_exists: bool,
 }
 
 fn default_web_terminal_port() -> u16 {
@@ -120,6 +162,7 @@ impl Default for AppSettings {
             default_microphone: None,
             dismissed_image_digest: None,
             web_terminal: WebTerminalSettings::default(),
+            stt: SttSettings::default(),
         }
     }
 }