Skip to content

Releases

CopySpeech v0.3.1

Choose a tag to compare

@github-actions github-actions released this 18 Nov 10:50

Added

  • tiny-voca engine – a new local engine with two voices (F2, M2). It runs on the CPU only, uses about 45 MB of files, and needs no API key. It reads English text only; other scripts fail with a clear message instead of silence. Windows installer only for now.

Changed

  • Windows starts on tiny-voca – fresh Windows installs use tiny-voca as the default engine. Existing settings keep their current engine. mac and Linux keep Kokoro.
  • Onboarding offers four engines – tiny-voca, Kokoro, Edge-TTS, and Fish Audio. tiny-voca appears first on Windows only. ElevenLabs and Cartesia left onboarding. Both stay in Settings.
  • Browser playback box – the floating playback control is now an icon-only pill that matches the DSH bar.

Fixed

  • Cloud engine options use selects – OpenAI response format, ElevenLabs output format and voice settings, and Cartesia output format and encoding now offer only the values the backend accepts. Free-text entries no longer reach the API.
  • Chrome companion connects again – installed builds registered the browser companion host with \\?\ extended-length paths, and Chrome fails the handshake for such paths, so the extension showed an error and double-copy fell back without the companion. The app now registers plain paths.

CopySpeech v0.3.0

Choose a tag to compare

@github-actions github-actions released this 16 Nov 22:19

Added

  • Chrome companion on Windows – installed builds register the browser companion host with Chrome when the app launches. Dev builds skip this. Edge, Firefox, Linux, and macOS are not covered yet.

Changed

  • CopySpeech – CopySpeak is now CopySpeech. Existing settings, history, and local engine files remain in their original folders.
  • Installer identity – CopySpeech may install alongside CopySpeak. Keep the old installation and its data until the upgrade path is verified. Close one version before starting the other; both use the same single-instance identity.

Fixed

  • Launch on login follows the upgrade – CopySpeech takes over the launch-on-login entry from CopySpeak, and removes the old entry, so one login starts one app.

CopySpeak v0.2.16

Choose a tag to compare

@github-actions github-actions released this 16 Nov 14:54

Added

  • CopySpeak for Hermes Desktop – read new assistant replies aloud and highlight spoken words. Choose full replies or excerpts, follow playback, and save plugin settings in Hermes.
  • Voice previews – browse voices in a bottom drawer and generate a preview when a provider sample is unavailable. Cloud previews can use provider credits.
  • CopySpeak DSH plugin β€” a new plugin reads assistant replies aloud through the CopySpeak desktop app. A sidebar panel turns auto-reading, captions, caption animation and follow-scroll on or off; a per-turn action reads the final reply of any turn; a Read selection pill reads just the selected words; and a floating bar pauses, seeks and stops. Spoken words paint live inside the reply, including inline code and code blocks.
  • Firefox companion β€” the browser companion now works on Firefox alongside Chrome.
  • Animations setting β€” a new Appearance toggle turns page fades, panel fades and smooth scrolling on or off. While the system asks for reduced motion, the switch reads as off, stays disabled, and animations stay off.

Changed

  • Engines dashboard – filter cloud and local engines. Test engines and manage installation from the same page.
  • Caption highlights – a gliding pill marks the current word on the Play page, HUD, and browser companion.
  • DSH partial auto-read – read the first paragraphs, the last paragraphs, or both for new replies.
  • Long readings pay as they play β€” paginated readings synthesize one fragment ahead of playback instead of the whole queue, so an early stop pays only for audio already heard, and pausing also pauses generation. Repeat clicks replay from a content-addressed speech cache (256 MiB, oldest-first eviction) even with history off, so re-reading a message costs nothing. Queued double-copy reads use the same just-in-time gate.
  • About page redesign β€” a new layout with smooth scrolling and thin scrollbars, acknowledgments as a single section with the footer merged in, and a creator credit line.
  • New landing hero demo β€” the hero plays an audio preview automatically with word-timed captions, and navigation links animate on hover and on the active page.
  • Darker app icons β€” the palette is rebuilt closer to the original art (a uniform 10% chroma trim) with a 20% black overlay baked in, landing 25% below the original chroma and 11 lightness points down, hue preserved. App, tray, installer and browser-extension icons all carry it.
  • OpenAI style presets β€” the free-text instructions field becomes a style preset dropdown.
  • Settings wording and options β€” the language selector gains an explicit System option, and the quick-settings double-copy toggle is labelled Listen in every language.

Fixed

  • Settings scrolling – growing sections update the scroll limit. Sidebar links can reach the full page.
  • DSH auto-read – opening a session no longer reads an old reply as a new one.
  • Readings no longer overlap β€” when two readings race, the second holds until the first finishes, across streaming and batch engines; stopping a queued reading removes it from the queue instead of letting it start later.
  • Playback controls hit the right reading β€” stop, pause and seek are scoped by reading id, and speaker ownership transfers at a reading's first audible sound, so stopping a reading that has not played yet leaves the audible one alone. The CLI rejects a --request-id with no value instead of sending an unscoped request.
  • Switching engines no longer double-loads models β€” the outgoing daemon retires before the replacement spawns, closing the CUDA out-of-memory window on profile switches; a wedged model load fails after a bounded 180 s wait instead of hanging.
  • DSH captions survive scrolling, rewinds and code blocks β€” highlights keep alignment across reply re-renders and cross-fragment rewinds, the highlight pill glides without text flashing, and words stay readable inside code chips.
  • Layout clamps β€” reading-view scroll, caption corner radius and profile panel resizing stay within bounds.

Removed

  • History export – removed the unused history export code and batch export controls.

CopySpeak v0.2.15

Choose a tag to compare

@github-actions github-actions released this 03 Nov 13:43
742f721

Added

  • xAI and Mistral cloud providers β€” xAI speaks with native word timings for highlighting and a 0.7–1.5Γ— speed range; Mistral uses Voxtral (voxtral-mini-tts-3-0). Both can refresh their voice lists and both work with Dramatize.
  • Post-process templates β€” a library of prompt templates with one active per profile. Built-ins are Dramatize, which adds emote tags and must keep every word, and Concise, which compresses for listening and may change words. Custom templates default to the rewrite class. Your old single prompt becomes a template the first time the new version loads it. The Home page has a Post-process switch with a template picker, disabled only when a word-preserving template meets an engine that cannot read emote tags.
  • Post-processing is set per profile β€” each profile has its own on/off switch and template. On upgrade, every profile inherits your previous global setting.
  • Stop from outside the app β€” POST /stop on the control server and copyspeak stop halt playback the same way the Stop button does. Hermes uses it to cut speech off when you interrupt, which also releases a blocked speak --wait.
  • Voice browser β€” search and filter voices by available metadata, listen to provider previews, and select with the keyboard. Voice lists load from a per-engine, per-key cache; Refresh fetches the latest list and shows provider errors instead of silently falling back. Long lists render one page at a time, so providers with thousands of voices stay fast to browse.
  • More onboarding choices β€” first-run setup now offers ElevenLabs and Fish Audio beside the existing options. Kokoro, ElevenLabs and Fish Audio are marked as recommended, Fish Audio is free through its included free model, and every card carries short copy translated into all nine languages.
  • Edge-TTS is back β€” Microsoft's free Read Aloud voices return on the fixed edge-tts 7.3.1 client, which moved to Microsoft's current host and handles the 403 clock-skew retry itself. Onboarding, the Engines page, About and the profile editor offer Edge again, and fresh installs get a keyless Edge fallback profile; existing configs can simply pick Edge.

Changed

  • Markdown bold and italic are kept by default β€” **bold**, __bold__, *italic* and _italic_ now reach the engine, which can voice them as emphasis. New configs start with stripping off; saved configs keep whatever they had, so turn it off in Settings if you want the new behavior. The Markdown syntax options are collapsed by default.
  • condense is now post_process β€” the control server field is post_process and the CLI flag is --post-process. The old condense field and --condense flag still work.
  • Piper installs become real Piper profiles β€” profiles the Piper installer created as generic local commands are converted to the Piper engine with its own options; hand-written local commands are left alone. Fresh installs write Piper profiles directly.
  • Gemini moves off the retiring preview model β€” profiles pinned to gemini-2.5-flash-preview-tts, which Google shuts down on 2026-11-17, or with no model set, switch to gemini-3.8-flash-tts. Custom model ids are kept.
  • Keys from .env stay out of the UI β€” the Engines page marks engines whose key comes from .env with a badge and shows a "from .env" placeholder instead of the key itself, and the key is never sent to the app window. Typing a key over one that comes from .env asks for confirmation, because the .env value keeps winning until you remove it and restart.
  • Settings reorganized β€” Voices is now Profiles; local engine idle shutdown and streaming moved to Advanced, and lowering other apps' audio moved to Playback and no longer requires Experimental. History and Post-processing appear only when they can take effect. Autosave, the global hotkey and the HUD now require Experimental; disabling it keeps their saved settings but turns them off.
  • Engine options collapse β€” in the profile editor, only the model stays expanded; the other engine settings (output format, speed, GPU acceleration, custom prompts) are collapsed until you open them.
  • Settings layout β€” Appearance is first, Import/Export follows Debug Mode, and each Acknowledgments provider is a single link spanning mark and name, highlighting and underlining together on hover. About also includes the creator credit, centered, and a Buy Me a Coffee link.
  • Language follows the system by default β€” the app chooses a supported system language when none is saved, including right-to-left layout for Hebrew and Arabic. Explicitly saved languages stay unchanged.
  • Shorter Settings copy β€” descriptions that repeated a row's tooltip are gone and labels are shorter ("History" instead of "Save Reading History").
  • Safer defaults for new configs β€” Lower Other Apps starts on, lowering other apps' audio to 10%, and the Effects switch and per-profile effects start off; your saved values are kept.
  • Speed up to 2Γ— β€” the speed slider now goes to 2Γ— instead of 1.5Γ—.
  • App appearance β€” stone-textured backgrounds, text-only top navigation, brighter light-mode form controls, and a subtle fade-in between pages replace the previous app styling. The content background gently shifts when scrolling, also on long pages, unless reduced motion is enabled.
  • Landing page refreshed β€” a new design includes an audio introduction with synchronized captions and a supported-engines carousel. The Listen demo ends when its audio finishes.
  • Edge requests share a process-wide cooldown β€” Retry-After is honored (capped at five minutes) or an escalating 5/10/20 s ladder capped at 60 s applies; request starts are spaced at least one second apart, and 403/429 refusals are never retried. Failures surface as safe status messages, not raw tracebacks.
  • Clipboard truncation warns wherever you are β€” the shortened-text notice is now a global toast, not a Play-page-only banner.
  • Kokoro streams while it synthesizes β€” the wrapper now yields one chunk per native inference window instead of joining the whole fragment, so a warm Kokoro daemon starts speaking after the first window instead of after the last. Captions arrive as a cumulative snapshot per window, the same way Piper streams per sentence (from the X-CopySpeak fork).
  • Changing the engine warms it immediately β€” switching profile in Settings, saving engine settings, or completing onboarding now prewarms the active engine in the background, so the first reading after a change no longer pays the full model load. Saving an unchanged config is a no-op (from the X-CopySpeak fork).

Fixed

  • Long xAI and Fish readings play again β€” readings long enough to split into several sections failed with "end of stream", because these providers send WAV headers with placeholder sizes. The app now corrects the header to the real length before decoding.
  • Experimental off really means off β€” with Experimental disabled, history no longer saves, caches or replays readings, retention cleanup no longer deletes, and ducking no longer starts, even when the hidden settings were saved as on. Saving or resetting during playback restores other apps' volume immediately.
  • Retrigger setting saves valid values β€” the menu offers Interrupt and Queue, the two values the app accepts; it previously offered options the app rejected.
  • Play never loses your edits β€” Play saves pending profile and settings edits before it starts speaking, a failed save is shown and stops the reading instead of speaking with stale settings, and a background config change no longer overwrites unsaved edits.
  • Profile import accepts every engine β€” imported profiles for Edge, Kitten, Qwen, Piper, Kokoro and Pocket were rejected as invalid; older exported profiles are also filled in with current defaults.
  • Effects switch reflects reality β€” a profile counts as having effects only when an effect is actually selected.
  • Fish voice refresh works with standard API keys β€” the browser lists licensed library voices instead of requesting account-owned models, which returned a permission error for standard keys.
  • Profile edits survive background reloads β€” external config changes no longer overwrite unsaved edits, and a save only marks the snapshot it sent as saved.
  • Tooltips respect right-to-left languages β€” Hebrew and Arabic tooltips keep their direction and wrap within the viewport.
  • Sharper Windows tray icon β€” the app sizes branded tray icons to the system's small-icon dimensions before handing them to Windows.
  • Windows app icon updates β€” changes to the app icon now trigger a resource rebuild, so an older icon no longer stays in the executable.
  • Gemini readings stick to the script β€” requests now anchor the model to read the text verbatim instead of treating clipboard content that reads like a question or command as a prompt, older preview models get the same direction inline where they ignore it, and a reply that already carries its own WAV header is no longer wrapped a second time.
  • Long readings start speaking sooner β€” Play-page, hotkey and control-server readings using non-streaming engines begin as each section is synthesized, while the complete reading remains available for replay and history.
  • Startup and tray preferences work together β€” Start in tray now applies on launch without flashing the main window; first-run setup and a missing tray still open normally. Hiding the tray disables hidden startup and close-to-tray behavior without discarding your preferences. Closing with Exit selected quits the whole app, including its background HUD. Launch-on-login registration errors now surface when saving settings instead of silently succeeding.
  • Predictable tray controls β€” left-click opens and restores CopySpeak on Windows without interrupting speech; right-click offers Open, Stop reading, Settings and Quit. Settings opens the Settings page. The b...
Read more

CopySpeak v0.2.14

Choose a tag to compare

@github-actions github-actions released this 20 Oct 04:26

Added

  • Fish Audio cloud provider β€” four models: s2.1-pro, s2.1-pro-free (the default, works on accounts without credits), s2-pro and drama-3. Playback starts while audio streams in. Fish returns no word timings, so word highlighting is estimated.

  • Dramatize (Experimental) β€” an LLM rewrites a reading with emote tags before it is spoken, for Home reads and double-copy. It needs an ElevenLabs v4 or v3 model, or a Fish S2 model. The Home switch is disabled, with the reason in a tooltip, when the active profile's model cannot read emote tags. A rewrite that adds, drops or changes any word is discarded, and the original text is spoken.

  • Experimental switch β€” in Advanced. It shows Dramatize, effects and Post-processing. Off, they are hidden and forced off; your saved settings are kept.

  • New app icon β€” the volume-glass mark, with a yellow tray icon while busy.

Changed

  • Settings sidebar β€” General, Playback, History, Advanced and About, plus Post-processing while Experimental is on. Each entry scrolls to its section. Every setting explains itself in a tooltip instead of text under the row.

  • Duck β€” on/off plus a level slider (default 20). The mode dropdown and fade slider are gone, and the fade is fixed at 150 ms. Users who had mode "mute" now duck to their level instead of going silent.

  • Effects β€” hidden unless Experimental is on, and the /effects page returns to Home while it is off. Saved effect settings are kept.

  • Post-processing β€” one Dramatize prompt replaces the three earlier prompts, and Restore default reads the backend default. Deleting a provider is removed; adding one stays.

  • Minimum length β€” new configs default to 0 characters. Existing configs keep their stored value.

Fixed

  • Browser companion follows double-copy captions in local file tabs β€” captions reach Markdown and other local files, not only web pages.

  • Browser companion wakes on a page copy β€” the first double-copy after the worker went idle no longer shows the stale ! badge.

  • Playback transport buttons in RTL β€” they rotate correctly in right-to-left layouts.

  • Updater in browser mode β€” skips Tauri-only calls instead of logging errors.

CopySpeak v0.2.13

Choose a tag to compare

@github-actions github-actions released this 16 Oct 04:11
6759936

Fixed

  • Browser companion stays connected on double-copy β€” the app could report the first word while a double-copied reading was still buffering, between the clip being queued and its audio starting. The extension rejects that, so it dropped its connection, showed ! on its icon and never highlighted the page. Words now reach the browser only while the reading plays or is paused.

CopySpeak v0.2.12

Choose a tag to compare

@github-actions github-actions released this 16 Oct 02:49
6e737d4

Added

  • Model picker for each engine β€” the model field for OpenAI, ElevenLabs, Cartesia and Google is now a list built from each vendor's documented models. A model id already saved in your config that isn't on the list stays selectable, so no setting is lost.

  • Tray menu follows the app language β€” the tray labels (the listening/paused toggle, Speak Clipboard, Settings, Quit) come from the same locale files as the window. Changing the language re-labels the tray without a restart.

  • Cyberpunk effect β€” a short synthesized 56k modem handshake opens and closes each reading, the first word and a few random later words glitch-stutter, and the voice gets a bright, lightly ring-modulated tone. Stutters add a little length, so karaoke captions drift slightly behind, as with the walkie-talkie lead-in.

Changed

  • OpenAI streams only on the gpt-4o-mini-tts family β€” the docs show chunked streaming for that model only. The default tts-1, tts-1-hd and unknown models now use the batch path, so default OpenAI users get their first audio only after the whole reading is synthesized. The docs don't confirm whether tts-1 can stream. To stream, switch to gpt-4o-mini-tts.

  • Local engines stream only when warm β€” kokoro and pocket always play a reading as one piece. piper, kitten and qwen stream only once their daemon is running; until then a reading takes the batch path.

  • More of the interface is translated β€” about 45 more strings now go through the translation catalog. Arabic, German, French, Hindi, Russian and Chinese have drafts for every key, and Spanish and Hebrew are extended to match. The drafts are unreviewed; reviewers check them in Crowdin.

  • ElevenLabs defaults to eleven_v4_turbo β€” fresh installs and the bundled ElevenLabs profile use the new model. A config that already saves eleven_v4 keeps it; change the model in the ElevenLabs engine settings.

Fixed

  • Failed streamed readings finish cleanly β€” a reading whose stream errored mid-way could leave playback stalled, with other audio ducked until the next reading. The failed reading now stops and finishes, and ducking restores (#70).

  • Aborts no longer reach the next reading β€” an abort applies only to readings already running. A late error or leftover audio from an aborted stream can't affect a reading that started after it.

  • ElevenLabs eleven_v4_turbo keeps its captions β€” the model sends audio-only records with no characters ahead of its text, and those cleared captions for the whole reading, so the live highlight never started. Those records now stay audio-only.

  • Russian and other Cyrillic text is no longer sent to the engine with some letters swapped for Latin lookalikes ("ΠΏΡ€ΠΈΠ²Π΅Ρ‚" became "ΠΏpΠΈΠ²eΡ‚"). Only words that mix Cyrillic and Latin letters, a common AI-text artifact, are still repaired.

  • Persian and Indic text keeps its zero-width non-joiner, so words spelled with it are no longer fused.

  • Hindi, Arabic, Urdu, Armenian, Amharic, Burmese and Khmer sentence endings now split long text between sentences instead of mid-word, and start a new caption phrase.

  • Playback events no longer fire twice β€” a page torn down before its event listeners finished registering left a listener behind, so every later event ran twice and a double-copy could speak twice.

CopySpeak v0.2.11

Choose a tag to compare

@github-actions github-actions released this 12 Oct 09:07

Removed

  • Edge-TTS is hidden everywhere in the UI β€” Microsoft killed the free Read-Aloud protocol the edge-tts CLI speaks (2026-10-10, rany2/edge-tts#490; every released edge-tts version 403s and no fixed release exists yet). Edge is gone from onboarding, the Engines page, the About page, and the profile editor's engine dropdown; existing Edge profiles are retired on config load (an active Edge profile switches to the bundled Kokoro profile). The engine code stays in the tree and comes back with a pin bump in src-tauri/src/tts/edge.rs once upstream ships a fixed release.

Added

  • Full transport controls on the Play page β€” rewind 5 s, play/pause, stop, and fast-forward 5 s buttons (Space toggles pause, ←/β†’ skip Β±5 s, Escape stops; ignored while typing). Skip works on both playback paths: streamed audio rebuilds its source timeline from a rolling 30 s history, and file-based playback seeks the media clock, stepping back into the previous fragment when a rewind crosses a fragment boundary. Stop now also aborts in-flight synthesis so local engines stop burning GPU/CPU immediately.

  • Spanish (es) interface language β€” selectable in Settings and on the landing page. Translations are managed in Crowdin (crowdin.yml); the Spanish text is a draft until reviewers approve it there.

  • Hebrew (he) interface language with right-to-left layout β€” selectable in Settings and on the landing page. The layout mirrors for RTL: text alignment, margins, and borders flip with the language. Translations are managed in Crowdin; the Hebrew text is a draft until reviewers approve it there.

  • Arabic, German, French, Hindi, Russian, and Simplified Chinese interface languages β€” ar, de, fr, hi, ru, and zh are selectable in Settings, each covering the full interface; Arabic uses the right-to-left layout. Translations are managed in Crowdin and are drafts until reviewers approve them there.

  • Browser companion follows double-copy and hotkey readings β€” with "Highlight when I double-copy" on (the default), the extension highlights the copied passage on the page as CopySpeak reads it, with Β±5 s skip on the pill. Each queued reading is followed only while its own audio plays.

  • LLM condensing controls β€” post-processing gains a minimum length (text shorter than it is spoken verbatim; default 200 characters), an "Only On Request" switch that condenses only when a caller asks (control server condense field or CLI --condense), and a request timeout so a slow provider falls back to the original text.

Fixed

  • Β±5 s skip lands where you are, not where the buffer started β€” skip targets were computed from a playhead that collapsed to 0 whenever synthesis lagged playback (stalls, the final drain, right after a skip), so a skip pressed near the end of a reading could replay it from the beginning. Fast-forward debt also consumed the wrong span at non-1Γ— speeds.
  • Global hotkey no longer fires twice β€” the shortcut plugin was registered twice with identical handlers, so one key press could start two readings.
  • A stopped queue no longer blocks later clipboard readings β€” stop_queue left a stale stop flag that silently no-op'd every subsequent double-copy / browser reading until restart.
  • Saved language applies before the first frame β€” the app painted in English (left to right) and then switched to the saved language after startup. The locale now loads before the first render.

Changed

  • Cleaner Play page β€” the "Paste text here, or copy it twice anywhere." hint and the 1 px separator lines (the sidebar divider and the horizontal rules around the quick settings) are gone; the quick settings now spread evenly down the sidebar instead of packing behind dividers.
  • Snappier playback edges β€” streamed readings now finish the moment their last audio drains instead of idling ~2 s before the HUD hides and blocking /speak --wait calls resolve, and first audio starts after ~150 ms of buffering instead of ~250 ms.
  • History is off by default for new installs β€” existing configs keep their saved setting. With history off, no reading is logged, and repeated readings of the same text still skip synthesis: the newest 15 clips are kept in CopySpeak's cache folder instead of the history folder. Readings made while history is off are no longer written to history.json, which is kept as-is.

CopySpeak v0.2.10

Choose a tag to compare

@github-actions github-actions released this 04 Oct 09:48

Changed

  • Cross-platform terminology β€” the Windows-only app identity is gone: general.start_with_windows is now general.launch_on_login (old config files migrate automatically; the autostart backend still matches each platform β€” registry Run key on Windows, XDG autostart on Linux, LaunchAgent on macOS), the settings toggle reads "Launch at login", and landing copy no longer claims Windows-only.

Added

  • macOS (Intel) port shipping β€” 0.2.9's macOS backend (clipboard trigger via NSPasteboard polling, selection capture, LaunchAgent autostart) is now the releasable 0.2.10 target.

CopySpeak v0.2.9

Choose a tag to compare

@github-actions github-actions released this 03 Oct 05:53

Added

  • Duck other apps while reading β€” new Settings β†’ Playback option lowers (or mutes) other apps' audio while CopySpeak speaks, then restores it. On by default, Windows only. Ducks per app, so CopySpeak itself stays audible; leaves any volume you change mid-read alone, ducks apps that start playing mid-read, and recovers the original volumes after a crash on next launch. Configurable level (default 20%), lower vs mute, and fade time.
  • Custom post-processing providers with live model lists β€” the LLM post-processing provider list is now a registry: Groq, OpenAI, Z.AI, OpenRouter, Ollama and Cerebras are seeded, and any OpenAI-compatible endpoint can be added, edited or removed. The model dropdown is filled live from the provider's /models endpoint, and keyless local endpoints (localhost) are allowed. Existing Groq settings migrate automatically.
  • Idle shutdown for local engines β€” resident local daemons (Kokoro, Qwen3, Kitten) now exit after a period without synthesis (default 10 minutes, set under Engines β†’ Idle shutdown; Never/1/5/10/30 min), freeing RAM/VRAM between reads. The next read cold-starts the engine.
  • Blocking /speak for Hermes β€” the control server's /speak accepts wait and returns once playback finishes; copyspeak --wait --ack writes a short silent WAV on success so command-provider integrations can treat it as a finished synthesis.

Fixed

  • /speak --wait no longer hangs until the 600 s cap β€” the playback-finished event now actually reaches the control server, so blocking calls end when speech ends or is stopped.

Changed

  • Removed the unused post-processing enum system and its dead UI; the post-processing credential check now validates the active provider.