From 4c10c73df07d8800b7b66414eb9c399075ef5080 Mon Sep 17 00:00:00 2001 From: oddlama Date: Sun, 26 Jan 2025 21:43:24 +0100 Subject: [PATCH] feat: add wyoming --- hosts/sausebiene/default.nix | 6 + hosts/sausebiene/home-assistant.nix | 4 +- hosts/sausebiene/pip.nix | 192 +++++++++++++++ hosts/sausebiene/wfw.nix | 346 ++++++++++++++++++++++++++++ hosts/sausebiene/wyoming.nix | 40 ++++ 5 files changed, 587 insertions(+), 1 deletion(-) create mode 100644 hosts/sausebiene/pip.nix create mode 100644 hosts/sausebiene/wfw.nix create mode 100644 hosts/sausebiene/wyoming.nix diff --git a/hosts/sausebiene/default.nix b/hosts/sausebiene/default.nix index e675672..6087f10 100644 --- a/hosts/sausebiene/default.nix +++ b/hosts/sausebiene/default.nix @@ -23,6 +23,7 @@ ./esphome.nix ./home-assistant.nix ./mosquitto.nix + ./wyoming.nix ]; topology.self.hardware.info = "Intel N100, 16GB RAM"; @@ -50,6 +51,11 @@ } ]; + systemd.tmpfiles.settings."01-var-lib-private"."/var/lib/private".d = { + user = "root"; + mode = "0700"; + }; + meta.promtail = { enable = true; proxy = "sentinel"; diff --git a/hosts/sausebiene/home-assistant.nix b/hosts/sausebiene/home-assistant.nix index 46d787c..845e91b 100644 --- a/hosts/sausebiene/home-assistant.nix +++ b/hosts/sausebiene/home-assistant.nix @@ -44,8 +44,10 @@ in "radio_browser" "soundtouch" # Bose SoundTouch "spotify" + "wake_word" "webostv" # LG WebOS TV - #"zha" + "whisper" + "wyoming" ]; customComponents = with pkgs.home-assistant-custom-components; [ diff --git a/hosts/sausebiene/pip.nix b/hosts/sausebiene/pip.nix new file mode 100644 index 0000000..90abefc --- /dev/null +++ b/hosts/sausebiene/pip.nix @@ -0,0 +1,192 @@ +{ + config, + lib, + pkgs, + utils, + ... +}: + +let + cfg = config.services.wyoming.piper; + + inherit (lib) + mkOption + mkEnableOption + mkPackageOption + types + ; + + inherit (builtins) + toString + ; + + inherit (utils) + escapeSystemdExecArgs + ; +in + +{ + options.services.wyoming.piper = with types; { + package = mkPackageOption pkgs "wyoming-piper" { }; + + servers = mkOption { + default = { }; + description = '' + Attribute set of wyoming-piper instances to spawn. + ''; + type = types.attrsOf ( + types.submodule (_: { + options = { + enable = mkEnableOption "Wyoming Piper server"; + + piper = mkPackageOption pkgs "piper-tts" { }; + + voice = mkOption { + type = str; + example = "en-us-ryan-medium"; + description = '' + Name of the voice model to use. See the following website for samples: + https://rhasspy.github.io/piper-samples/ + ''; + }; + + uri = mkOption { + type = strMatching "^(tcp|unix)://.*$"; + example = "tcp://0.0.0.0:10200"; + description = '' + URI to bind the wyoming server to. + ''; + }; + + speaker = mkOption { + type = ints.unsigned; + default = 0; + description = '' + ID of a specific speaker in a multi-speaker model. + ''; + apply = toString; + }; + + noiseScale = mkOption { + type = numbers.between 0.0 1.0; + default = 0.667; + description = '' + Generator noise value. + ''; + apply = toString; + }; + + noiseWidth = mkOption { + type = numbers.between 0.0 1.0; + default = 0.333; + description = '' + Phoneme width noise value. + ''; + apply = toString; + }; + + lengthScale = mkOption { + type = numbers.between 0.0 1.0; + default = 1.0; + description = '' + Phoneme length value. + ''; + apply = toString; + }; + + extraArgs = mkOption { + type = listOf str; + default = [ ]; + description = '' + Extra arguments to pass to the server commandline. + ''; + }; + }; + }) + ); + }; + }; + + config = + let + inherit (lib) + mapAttrs' + mkIf + nameValuePair + ; + in + mkIf (cfg.servers != { }) { + systemd.services = mapAttrs' ( + server: options: + nameValuePair "wyoming-piper-${server}" { + inherit (options) enable; + description = "Wyoming Piper server instance ${server}"; + wants = [ + "network-online.target" + ]; + after = [ + "network-online.target" + ]; + wantedBy = [ + "multi-user.target" + ]; + serviceConfig = { + DynamicUser = true; + User = "wyoming-piper"; + StateDirectory = [ "wyoming/piper" ]; + # https://github.com/home-assistant/addons/blob/master/piper/rootfs/etc/s6-overlay/s6-rc.d/piper/run + ExecStart = escapeSystemdExecArgs ( + [ + (lib.getExe cfg.package) + "--data-dir" + "/var/lib/wyoming/piper" + "--uri" + options.uri + "--piper" + (lib.getExe options.piper) + "--voice" + options.voice + "--speaker" + options.speaker + "--length-scale" + options.lengthScale + "--noise-scale" + options.noiseScale + "--noise-w" + options.noiseWidth + ] + ++ options.extraArgs + ); + CapabilityBoundingSet = ""; + DeviceAllow = ""; + DevicePolicy = "closed"; + LockPersonality = true; + MemoryDenyWriteExecute = true; + PrivateDevices = true; + PrivateUsers = true; + ProtectHome = true; + ProtectHostname = true; + ProtectKernelLogs = true; + ProtectKernelModules = true; + ProtectKernelTunables = true; + ProtectControlGroups = true; + ProtectProc = "invisible"; + ProcSubset = "pid"; + RestrictAddressFamilies = [ + "AF_INET" + "AF_INET6" + "AF_UNIX" + ]; + RestrictNamespaces = true; + RestrictRealtime = true; + SystemCallArchitectures = "native"; + SystemCallFilter = [ + "@system-service" + "~@privileged" + ]; + UMask = "0077"; + }; + } + ) cfg.servers; + }; +} diff --git a/hosts/sausebiene/wfw.nix b/hosts/sausebiene/wfw.nix new file mode 100644 index 0000000..a79cf04 --- /dev/null +++ b/hosts/sausebiene/wfw.nix @@ -0,0 +1,346 @@ +{ + config, + lib, + pkgs, + utils, + ... +}: + +let + cfg = config.services.wyoming.faster-whisper; + + inherit (lib) + mkOption + mkEnableOption + mkPackageOption + types + ; + + inherit (builtins) + toString + ; + + inherit (utils) + escapeSystemdExecArgs + ; + +in + +{ + options.services.wyoming.faster-whisper = with types; { + package = mkPackageOption pkgs "wyoming-faster-whisper" { }; + + servers = mkOption { + default = { }; + description = '' + Attribute set of wyoming-faster-whisper instances to spawn. + ''; + type = types.attrsOf ( + types.submodule (_: { + options = { + enable = mkEnableOption "Wyoming faster-whisper server"; + + model = mkOption { + type = str; + default = "tiny-int8"; + example = "Systran/faster-distil-whisper-small.en"; + description = '' + Name of the voice model to use. Can also be a HuggingFace model ID or a path to + a custom model directory. + + Compressed models (`int8`) are slightly less accurate, but smaller and faster. + + Available models: + - `tiny-int8` (compressed) + - `tiny` + - `tiny.en` (English only) + - `base-int8` (compressed) + - `base` + - `base.en` (English only) + - `small-int8` (compressed) + - `distil-small.en` (distilled, English only) + - `small` + - `small.en` (English only) + - `medium-int8` (compressed) + - `distil-medium.en` (distilled, English only) + - `medium` + - `medium.en` (English only) + - `large` + - `large-v1` + - `distil-large-v2` (distilled, English only) + - `large-v2` + - `distil-large-v3` (distilled, English only) + - `large-v3` + - `turbo` (faster than large-v3) + ''; + }; + + uri = mkOption { + type = strMatching "^(tcp|unix)://.*$"; + example = "tcp://0.0.0.0:10300"; + description = '' + URI to bind the wyoming server to. + ''; + }; + + device = mkOption { + # https://opennmt.net/CTranslate2/python/ctranslate2.models.Whisper.html# + type = types.enum [ + "cpu" + "cuda" + "auto" + ]; + default = "cpu"; + description = '' + Determines the platform faster-whisper is run on. CPU works everywhere, CUDA requires a compatible NVIDIA GPU. + ''; + }; + + language = mkOption { + type = enum [ + # https://github.com/home-assistant/addons/blob/master/whisper/config.yaml#L20 + "auto" + "af" + "am" + "ar" + "as" + "az" + "ba" + "be" + "bg" + "bn" + "bo" + "br" + "bs" + "ca" + "cs" + "cy" + "da" + "de" + "el" + "en" + "es" + "et" + "eu" + "fa" + "fi" + "fo" + "fr" + "gl" + "gu" + "ha" + "haw" + "he" + "hi" + "hr" + "ht" + "hu" + "hy" + "id" + "is" + "it" + "ja" + "jw" + "ka" + "kk" + "km" + "kn" + "ko" + "la" + "lb" + "ln" + "lo" + "lt" + "lv" + "mg" + "mi" + "mk" + "ml" + "mn" + "mr" + "ms" + "mt" + "my" + "ne" + "nl" + "nn" + "no" + "oc" + "pa" + "pl" + "ps" + "pt" + "ro" + "ru" + "sa" + "sd" + "si" + "sk" + "sl" + "sn" + "so" + "sq" + "sr" + "su" + "sv" + "sw" + "ta" + "te" + "tg" + "th" + "tk" + "tl" + "tr" + "tt" + "uk" + "ur" + "uz" + "vi" + "yi" + "yue" + "yo" + "zh" + ]; + example = "en"; + description = '' + The language used to to parse words and sentences. + ''; + }; + + initialPrompt = mkOption { + type = nullOr str; + default = null; + example = '' + The following conversation takes place in the universe of Wizard of Oz. Key terms include 'Yellow Brick Road' (the path to follow), 'Emerald City' (the ultimate goal), and 'Ruby Slippers' (the magical tools to succeed). Keep these in mind as they guide the journey. + ''; + description = '' + Optional text to provide as a prompt for the first window. This can be used to provide, or + "prompt-engineer" a context for transcription, e.g. custom vocabularies or proper nouns + to make it more likely to predict those word correctly. + ''; + }; + + beamSize = mkOption { + type = ints.unsigned; + default = 0; + example = 5; + description = '' + The number of beams to use in beam search. + Use `0` to automatically select a value based on the CPU. + ''; + apply = toString; + }; + + extraArgs = mkOption { + type = listOf str; + default = [ ]; + description = '' + Extra arguments to pass to the server commandline. + ''; + }; + }; + }) + ); + }; + }; + + config = + let + inherit (lib) + mapAttrs' + mkIf + nameValuePair + ; + in + mkIf (cfg.servers != { }) { + systemd.services = mapAttrs' ( + server: options: + nameValuePair "wyoming-faster-whisper-${server}" { + inherit (options) enable; + description = "Wyoming faster-whisper server instance ${server}"; + wants = [ + "network-online.target" + ]; + after = [ + "network-online.target" + ]; + wantedBy = [ + "multi-user.target" + ]; + # https://github.com/rhasspy/wyoming-faster-whisper/issues/27 + environment."HF_HUB_CACHE" = "/tmp"; + serviceConfig = { + DynamicUser = true; + User = "wyoming-faster-whisper"; + StateDirectory = [ "wyoming/faster-whisper" ]; + # https://github.com/home-assistant/addons/blob/master/whisper/rootfs/etc/s6-overlay/s6-rc.d/whisper/run + ExecStart = escapeSystemdExecArgs ( + [ + (lib.getExe cfg.package) + "--data-dir" + "/var/lib/wyoming/faster-whisper" + "--uri" + options.uri + "--device" + options.device + "--model" + options.model + "--language" + options.language + "--beam-size" + options.beamSize + ] + ++ lib.optionals (options.initialPrompt != null) [ + "--initial-prompt" + options.initialPrompt + ] + ++ options.extraArgs + ); + CapabilityBoundingSet = ""; + DeviceAllow = + if + builtins.elem options.device [ + "cuda" + "auto" + ] + then + [ + # https://docs.nvidia.com/dgx/pdf/dgx-os-5-user-guide.pdf + "char-nvidia-uvm" + "char-nvidia-frontend" + "char-nvidia-caps" + "char-nvidiactl" + ] + else + ""; + DevicePolicy = "closed"; + LockPersonality = true; + MemoryDenyWriteExecute = true; + PrivateUsers = true; + ProtectHome = true; + ProtectHostname = true; + ProtectKernelLogs = true; + ProtectKernelModules = true; + ProtectKernelTunables = true; + ProtectControlGroups = true; + ProtectProc = "invisible"; + # "all" is required because faster-whisper accesses /proc/cpuinfo to determine cpu capabilities + ProcSubset = "all"; + RestrictAddressFamilies = [ + "AF_INET" + "AF_INET6" + "AF_UNIX" + ]; + RestrictNamespaces = true; + RestrictRealtime = true; + SystemCallArchitectures = "native"; + SystemCallFilter = [ + "@system-service" + "~@privileged" + ]; + UMask = "0077"; + }; + } + ) cfg.servers; + }; +} diff --git a/hosts/sausebiene/wyoming.nix b/hosts/sausebiene/wyoming.nix new file mode 100644 index 0000000..54aca81 --- /dev/null +++ b/hosts/sausebiene/wyoming.nix @@ -0,0 +1,40 @@ +{ + disabledModules = [ + "services/home-automation/wyoming/faster-whisper.nix" + "services/home-automation/wyoming/piper.nix" + ]; + imports = [ + (builtins.trace "remove after next flake update" ./wfw.nix) + (builtins.trace "remove after next flake update" ./pip.nix) + ]; + + environment.persistence."/persist".directories = [ + { + directory = "/var/lib/private/wyoming"; + mode = "0700"; + } + ]; + + services.wyoming.faster-whisper = { + servers.hass = { + enable = true; + # see https://github.com/rhasspy/rhasspy3/blob/master/programs/asr/faster-whisper/script/download.py + model = "base-int8"; + language = "de"; + uri = "tcp://0.0.0.0:10300"; + device = "cpu"; + }; + }; + + services.wyoming.piper = { + servers.hass = { + enable = true; + # https://rhasspy.github.io/piper-samples/ + voice = "de_DE-thorsten-high"; + uri = "tcp://0.0.0.0:10200"; + }; + }; + + # needs access to /proc/cpuinfo + # systemd.services."wyoming-faster-whisper-en".serviceConfig.ProcSubset = lib.mkForce "all"; +}