speechflow 0.9.8 → 1.0.0

package/dst/speechflow-node-deepgram.js

@@ -1,220 +0,0 @@
-"use strict";
-/*
-**  SpeechFlow - Speech Processing Flow Graph
-**  Copyright (c) 2024-2025 Dr. Ralf S. Engelschall <rse@engelschall.com>
-**  Licensed under GPL 3.0 <https://spdx.org/licenses/GPL-3.0-only>
-*/
-var __createBinding = (this && this.__createBinding) || (Object.create ? (function(o, m, k, k2) {
-    if (k2 === undefined) k2 = k;
-    var desc = Object.getOwnPropertyDescriptor(m, k);
-    if (!desc || ("get" in desc ? !m.__esModule : desc.writable || desc.configurable)) {
-      desc = { enumerable: true, get: function() { return m[k]; } };
-    }
-    Object.defineProperty(o, k2, desc);
-}) : (function(o, m, k, k2) {
-    if (k2 === undefined) k2 = k;
-    o[k2] = m[k];
-}));
-var __setModuleDefault = (this && this.__setModuleDefault) || (Object.create ? (function(o, v) {
-    Object.defineProperty(o, "default", { enumerable: true, value: v });
-}) : function(o, v) {
-    o["default"] = v;
-});
-var __importStar = (this && this.__importStar) || (function () {
-    var ownKeys = function(o) {
-        ownKeys = Object.getOwnPropertyNames || function (o) {
-            var ar = [];
-            for (var k in o) if (Object.prototype.hasOwnProperty.call(o, k)) ar[ar.length] = k;
-            return ar;
-        };
-        return ownKeys(o);
-    };
-    return function (mod) {
-        if (mod && mod.__esModule) return mod;
-        var result = {};
-        if (mod != null) for (var k = ownKeys(mod), i = 0; i < k.length; i++) if (k[i] !== "default") __createBinding(result, mod, k[i]);
-        __setModuleDefault(result, mod);
-        return result;
-    };
-})();
-var __importDefault = (this && this.__importDefault) || function (mod) {
-    return (mod && mod.__esModule) ? mod : { "default": mod };
-};
-Object.defineProperty(exports, "__esModule", { value: true });
-/*  standard dependencies  */
-const node_stream_1 = __importDefault(require("node:stream"));
-/*  external dependencies  */
-const Deepgram = __importStar(require("@deepgram/sdk"));
-const luxon_1 = require("luxon");
-/*  internal dependencies  */
-const speechflow_node_1 = __importStar(require("./speechflow-node"));
-const utils = __importStar(require("./speechflow-utils"));
-/*  SpeechFlow node for Deepgram speech-to-text conversion  */
-class SpeechFlowNodeDeepgram extends speechflow_node_1.default {
-    /*  declare official node name  */
-    static name = "deepgram";
-    /*  internal state  */
-    dg = null;
-    /*  construct node  */
-    constructor(id, cfg, opts, args) {
-        super(id, cfg, opts, args);
-        /*  declare node configuration parameters  */
-        this.configure({
-            key: { type: "string", val: process.env.SPEECHFLOW_KEY_DEEPGRAM },
-            model: { type: "string", val: "nova-3", pos: 0 },
-            version: { type: "string", val: "latest", pos: 1 },
-            language: { type: "string", val: "multi", pos: 2 }
-        });
-        /*  declare node input/output format  */
-        this.input = "audio";
-        this.output = "text";
-    }
-    /*  open node  */
-    async open() {
-        /*  sanity check situation  */
-        if (this.config.audioBitDepth !== 16 || !this.config.audioLittleEndian)
-            throw new Error("Deepgram node currently supports PCM-S16LE audio only");
-        /*  create queue for results  */
-        const queue = new utils.SingleQueue();
-        /*  connect to Deepgram API  */
-        const deepgram = Deepgram.createClient(this.params.key);
-        let language = "en";
-        if (this.params.model.match(/^nova-2/) && this.params.language !== "en")
-            language = this.params.language;
-        else if (this.params.model.match(/^nova-3/) && this.params.language !== "en")
-            language = "multi";
-        this.dg = deepgram.listen.live({
-            mip_opt_out: true,
-            model: this.params.model,
-            version: this.params.version,
-            language,
-            channels: this.config.audioChannels,
-            sample_rate: this.config.audioSampleRate,
-            encoding: "linear16",
-            multichannel: false,
-            endpointing: 10,
-            interim_results: false,
-            smart_format: true,
-            punctuate: true,
-            filler_words: true,
-            diarize: true, /* still not used by us */
-            numerals: true,
-            profanity_filter: false
-        });
-        /*  hook onto Deepgram API events  */
-        this.dg.on(Deepgram.LiveTranscriptionEvents.Transcript, async (data) => {
-            const text = data.channel?.alternatives[0].transcript ?? "";
-            if (text === "")
-                this.log("info", `Deepgram: empty/dummy text received (start: ${data.start}s, duration: ${data.duration}s)`);
-            else {
-                this.log("info", `Deepgram: text received (start: ${data.start}s, duration: ${data.duration}s): "${text}"`);
-                const start = luxon_1.Duration.fromMillis(data.start * 1000).plus(this.timeZeroOffset);
-                const end = start.plus({ seconds: data.duration });
-                const chunk = new speechflow_node_1.SpeechFlowChunk(start, end, "final", "text", text);
-                queue.write(chunk);
-            }
-        });
-        this.dg.on(Deepgram.LiveTranscriptionEvents.Metadata, (data) => {
-            this.log("info", "Deepgram: metadata received");
-        });
-        this.dg.on(Deepgram.LiveTranscriptionEvents.Close, () => {
-            this.log("info", "Deepgram: connection close");
-        });
-        this.dg.on(Deepgram.LiveTranscriptionEvents.Error, (error) => {
-            this.log("error", `Deepgram: ${error.message}`);
-            this.emit("error");
-        });
-        /*  wait for Deepgram API to be available  */
-        await new Promise((resolve, reject) => {
-            let timer = setTimeout(() => {
-                if (timer !== null) {
-                    timer = null;
-                    reject(new Error("Deepgram: timeout waiting for connection open"));
-                }
-            }, 3000);
-            this.dg.once(Deepgram.LiveTranscriptionEvents.Open, () => {
-                this.log("info", "Deepgram: connection open");
-                if (timer !== null) {
-                    clearTimeout(timer);
-                    timer = null;
-                }
-                resolve(true);
-            });
-        });
-        /*  remember opening time to receive time zero offset  */
-        this.timeOpen = luxon_1.DateTime.now();
-        /*  workaround Deepgram initialization problems  */
-        let initDone = false;
-        let initTimeout = null;
-        const initTimeoutStart = () => {
-            if (initDone)
-                return;
-            setTimeout(async () => {
-                if (initTimeout === null)
-                    return;
-                initTimeout = null;
-                this.log("warning", "Deepgram: initialization timeout -- restarting service usage");
-                await this.close();
-                this.open();
-            }, 3000);
-        };
-        const initTimeoutStop = () => {
-            if (initDone)
-                return;
-            initDone = true;
-            if (initTimeout !== null) {
-                clearTimeout(initTimeout);
-                initTimeout = null;
-            }
-        };
-        /*  provide Duplex stream and internally attach to Deepgram API  */
-        const dg = this.dg;
-        const log = (level, msg) => {
-            this.log(level, msg);
-        };
-        const encoding = this.config.textEncoding;
-        this.stream = new node_stream_1.default.Duplex({
-            writableObjectMode: true,
-            readableObjectMode: true,
-            decodeStrings: false,
-            write(chunk, encoding, callback) {
-                if (chunk.type !== "audio")
-                    callback(new Error("expected audio input chunk"));
-                else if (!Buffer.isBuffer(chunk.payload))
-                    callback(new Error("expected Buffer input chunk"));
-                else {
-                    if (chunk.payload.byteLength > 0) {
-                        log("info", `Deepgram: send data (${chunk.payload.byteLength} bytes)`);
-                        initTimeoutStart();
-                        dg.send(chunk.payload); /* intentionally discard all time information  */
-                    }
-                    callback();
-                }
-            },
-            read(size) {
-                queue.read().then((chunk) => {
-                    log("info", `Deepgram: receive data (${chunk.payload.length} bytes)`);
-                    initTimeoutStop();
-                    this.push(chunk, encoding);
-                });
-            },
-            final(callback) {
-                dg.requestClose();
-                this.push(null);
-                callback();
-            }
-        });
-    }
-    /*  close node  */
-    async close() {
-        /*  close stream  */
-        if (this.stream !== null) {
-            this.stream.destroy();
-            this.stream = null;
-        }
-        /*  shutdown Deepgram API  */
-        if (this.dg !== null)
-            this.dg.requestClose();
-    }
-}
-exports.default = SpeechFlowNodeDeepgram;

package/dst/speechflow-node-deepl.js

@@ -1,128 +0,0 @@
-"use strict";
-/*
-**  SpeechFlow - Speech Processing Flow Graph
-**  Copyright (c) 2024-2025 Dr. Ralf S. Engelschall <rse@engelschall.com>
-**  Licensed under GPL 3.0 <https://spdx.org/licenses/GPL-3.0-only>
-*/
-var __createBinding = (this && this.__createBinding) || (Object.create ? (function(o, m, k, k2) {
-    if (k2 === undefined) k2 = k;
-    var desc = Object.getOwnPropertyDescriptor(m, k);
-    if (!desc || ("get" in desc ? !m.__esModule : desc.writable || desc.configurable)) {
-      desc = { enumerable: true, get: function() { return m[k]; } };
-    }
-    Object.defineProperty(o, k2, desc);
-}) : (function(o, m, k, k2) {
-    if (k2 === undefined) k2 = k;
-    o[k2] = m[k];
-}));
-var __setModuleDefault = (this && this.__setModuleDefault) || (Object.create ? (function(o, v) {
-    Object.defineProperty(o, "default", { enumerable: true, value: v });
-}) : function(o, v) {
-    o["default"] = v;
-});
-var __importStar = (this && this.__importStar) || (function () {
-    var ownKeys = function(o) {
-        ownKeys = Object.getOwnPropertyNames || function (o) {
-            var ar = [];
-            for (var k in o) if (Object.prototype.hasOwnProperty.call(o, k)) ar[ar.length] = k;
-            return ar;
-        };
-        return ownKeys(o);
-    };
-    return function (mod) {
-        if (mod && mod.__esModule) return mod;
-        var result = {};
-        if (mod != null) for (var k = ownKeys(mod), i = 0; i < k.length; i++) if (k[i] !== "default") __createBinding(result, mod, k[i]);
-        __setModuleDefault(result, mod);
-        return result;
-    };
-})();
-var __importDefault = (this && this.__importDefault) || function (mod) {
-    return (mod && mod.__esModule) ? mod : { "default": mod };
-};
-Object.defineProperty(exports, "__esModule", { value: true });
-/*  standard dependencies  */
-const node_stream_1 = __importDefault(require("node:stream"));
-/*  external dependencies  */
-const DeepL = __importStar(require("deepl-node"));
-/*  internal dependencies  */
-const speechflow_node_1 = __importDefault(require("./speechflow-node"));
-/*  SpeechFlow node for DeepL text-to-text translations  */
-class SpeechFlowNodeDeepL extends speechflow_node_1.default {
-    /*  declare official node name  */
-    static name = "deepl";
-    /*  internal state  */
-    deepl = null;
-    /*  construct node  */
-    constructor(id, cfg, opts, args) {
-        super(id, cfg, opts, args);
-        /*  declare node configuration parameters  */
-        this.configure({
-            key: { type: "string", val: process.env.SPEECHFLOW_KEY_DEEPL },
-            src: { type: "string", pos: 0, val: "de", match: /^(?:de|en-US)$/ },
-            dst: { type: "string", pos: 1, val: "en-US", match: /^(?:de|en-US)$/ },
-            optimize: { type: "string", pos: 2, val: "latency", match: /^(?:latency|quality)$/ }
-        });
-        /*  declare node input/output format  */
-        this.input = "text";
-        this.output = "text";
-    }
-    /*  open node  */
-    async open() {
-        /*  instantiate DeepL API SDK  */
-        this.deepl = new DeepL.Translator(this.params.key);
-        /*  provide text-to-text translation  */
-        const translate = async (text) => {
-            const result = await this.deepl.translateText(text, this.params.src, this.params.dst, {
-                splitSentences: "off",
-                modelType: this.params.optimize === "latency" ?
-                    "latency_optimized" : "prefer_quality_optimized",
-                preserveFormatting: true,
-                formality: "prefer_more"
-            });
-            return (result?.text ?? text);
-        };
-        /*  establish a duplex stream and connect it to DeepL translation  */
-        this.stream = new node_stream_1.default.Transform({
-            readableObjectMode: true,
-            writableObjectMode: true,
-            decodeStrings: false,
-            transform(chunk, encoding, callback) {
-                if (Buffer.isBuffer(chunk.payload))
-                    callback(new Error("invalid chunk payload type"));
-                else {
-                    if (chunk.payload === "") {
-                        this.push(chunk);
-                        callback();
-                    }
-                    else {
-                        translate(chunk.payload).then((payload) => {
-                            const chunkNew = chunk.clone();
-                            chunkNew.payload = payload;
-                            this.push(chunkNew);
-                            callback();
-                        }).catch((err) => {
-                            callback(err);
-                        });
-                    }
-                }
-            },
-            final(callback) {
-                this.push(null);
-                callback();
-            }
-        });
-    }
-    /*  open node  */
-    async close() {
-        /*  close stream  */
-        if (this.stream !== null) {
-            this.stream.destroy();
-            this.stream = null;
-        }
-        /*  shutdown DeepL API  */
-        if (this.deepl !== null)
-            this.deepl = null;
-    }
-}
-exports.default = SpeechFlowNodeDeepL;

package/dst/speechflow-node-device.d.ts

@@ -1,13 +0,0 @@
-import SpeechFlowNode from "./speechflow-node";
-export default class SpeechFlowNodeDevice extends SpeechFlowNode {
-    static name: string;
-    private io;
-    constructor(id: string, cfg: {
-        [id: string]: any;
-    }, opts: {
-        [id: string]: any;
-    }, args: any[]);
-    private audioDeviceFromURL;
-    open(): Promise<void>;
-    close(): Promise<void>;
-}

package/dst/speechflow-node-device.js

@@ -1,205 +0,0 @@
-"use strict";
-/*
-**  SpeechFlow - Speech Processing Flow Graph
-**  Copyright (c) 2024-2025 Dr. Ralf S. Engelschall <rse@engelschall.com>
-**  Licensed under GPL 3.0 <https://spdx.org/licenses/GPL-3.0-only>
-*/
-var __createBinding = (this && this.__createBinding) || (Object.create ? (function(o, m, k, k2) {
-    if (k2 === undefined) k2 = k;
-    var desc = Object.getOwnPropertyDescriptor(m, k);
-    if (!desc || ("get" in desc ? !m.__esModule : desc.writable || desc.configurable)) {
-      desc = { enumerable: true, get: function() { return m[k]; } };
-    }
-    Object.defineProperty(o, k2, desc);
-}) : (function(o, m, k, k2) {
-    if (k2 === undefined) k2 = k;
-    o[k2] = m[k];
-}));
-var __setModuleDefault = (this && this.__setModuleDefault) || (Object.create ? (function(o, v) {
-    Object.defineProperty(o, "default", { enumerable: true, value: v });
-}) : function(o, v) {
-    o["default"] = v;
-});
-var __importStar = (this && this.__importStar) || (function () {
-    var ownKeys = function(o) {
-        ownKeys = Object.getOwnPropertyNames || function (o) {
-            var ar = [];
-            for (var k in o) if (Object.prototype.hasOwnProperty.call(o, k)) ar[ar.length] = k;
-            return ar;
-        };
-        return ownKeys(o);
-    };
-    return function (mod) {
-        if (mod && mod.__esModule) return mod;
-        var result = {};
-        if (mod != null) for (var k = ownKeys(mod), i = 0; i < k.length; i++) if (k[i] !== "default") __createBinding(result, mod, k[i]);
-        __setModuleDefault(result, mod);
-        return result;
-    };
-})();
-var __importDefault = (this && this.__importDefault) || function (mod) {
-    return (mod && mod.__esModule) ? mod : { "default": mod };
-};
-Object.defineProperty(exports, "__esModule", { value: true });
-/*  standard dependencies  */
-const node_stream_1 = __importDefault(require("node:stream"));
-/*  external dependencies  */
-const naudiodon_1 = __importDefault(require("@gpeng/naudiodon"));
-/*  internal dependencies  */
-const speechflow_node_1 = __importDefault(require("./speechflow-node"));
-const utils = __importStar(require("./speechflow-utils"));
-/*  SpeechFlow node for device access  */
-class SpeechFlowNodeDevice extends speechflow_node_1.default {
-    /*  declare official node name  */
-    static name = "device";
-    /*  internal state  */
-    io = null;
-    /*  construct node  */
-    constructor(id, cfg, opts, args) {
-        super(id, cfg, opts, args);
-        /*  declare node configuration parameters  */
-        this.configure({
-            device: { type: "string", pos: 0, match: /^(.+?):(.+)$/ },
-            mode: { type: "string", pos: 1, val: "rw", match: /^(?:r|w|rw)$/ }
-        });
-        /*  declare node input/output format  */
-        if (this.params.mode === "rw") {
-            this.input = "audio";
-            this.output = "audio";
-        }
-        else if (this.params.mode === "r") {
-            this.input = "none";
-            this.output = "audio";
-        }
-        else if (this.params.mode === "w") {
-            this.input = "audio";
-            this.output = "none";
-        }
-    }
-    /*  INTERNAL: utility function for finding audio device by pseudo-URL notation  */
-    audioDeviceFromURL(mode, url) {
-        /*  parse URL  */
-        const m = url.match(/^(.+?):(.+)$/);
-        if (m === null)
-            throw new Error(`invalid audio device URL "${url}"`);
-        const [, type, name] = m;
-        /*  determine audio API  */
-        const apis = naudiodon_1.default.getHostAPIs();
-        const api = apis.HostAPIs.find((api) => api.type.toLowerCase() === type.toLowerCase());
-        if (!api)
-            throw new Error(`invalid audio API type "${type}"`);
-        /*  determine device of audio API  */
-        const devices = naudiodon_1.default.getDevices();
-        for (const device of devices)
-            this.log("info", `found audio device "${device.name}" ` +
-                `(inputs: ${device.maxInputChannels}, outputs: ${device.maxOutputChannels}`);
-        const device = devices.find((device) => {
-            return (((mode === "r" && device.maxInputChannels > 0)
-                || (mode === "w" && device.maxOutputChannels > 0)
-                || (mode === "rw" && device.maxInputChannels > 0 && device.maxOutputChannels > 0)
-                || (mode === "any" && (device.maxInputChannels > 0 || device.maxOutputChannels > 0)))
-                && device.name.match(name)
-                && device.hostAPIName === api.name);
-        });
-        if (!device)
-            throw new Error(`invalid audio device "${name}" (of audio API type "${type}")`);
-        return device;
-    }
-    /*  open node  */
-    async open() {
-        /*  determine device  */
-        const device = this.audioDeviceFromURL(this.params.mode, this.params.device);
-        /*  sanity check sample rate compatibility
-            (we still do not resample in input/output for simplification reasons)  */
-        if (device.defaultSampleRate !== this.config.audioSampleRate)
-            throw new Error(`audio device sample rate ${device.defaultSampleRate} is ` +
-                `incompatible with required sample rate ${this.config.audioSampleRate}`);
-        /*  establish device connection
-            Notice: "naudion" actually implements Stream.{Readable,Writable,Duplex}, but
-            declares just its sub-interface NodeJS.{Readable,Writable,Duplex}Stream,
-            so it is correct to cast it back to Stream.{Readable,Writable,Duplex}  */
-        /*  FIXME: the underlying PortAudio outputs verbose/debugging messages  */
-        if (this.params.mode === "rw") {
-            /*  input/output device  */
-            if (device.maxInputChannels === 0)
-                throw new Error(`device "${device.id}" does not have any input channels (required by read/write mode)`);
-            if (device.maxOutputChannels === 0)
-                throw new Error(`device "${device.id}" does not have any output channels (required by read/write mode)`);
-            this.log("info", `resolved "${this.params.device}" to duplex device "${device.id}"`);
-            this.io = naudiodon_1.default.AudioIO({
-                inOptions: {
-                    deviceId: device.id,
-                    channelCount: this.config.audioChannels,
-                    sampleRate: this.config.audioSampleRate,
-                    sampleFormat: this.config.audioBitDepth
-                },
-                outOptions: {
-                    deviceId: device.id,
-                    channelCount: this.config.audioChannels,
-                    sampleRate: this.config.audioSampleRate,
-                    sampleFormat: this.config.audioBitDepth
-                }
-            });
-            this.stream = this.io;
-            /*  convert regular stream into object-mode stream  */
-            const wrapper1 = utils.createTransformStreamForWritableSide();
-            const wrapper2 = utils.createTransformStreamForReadableSide("audio", () => this.timeZero);
-            this.stream = node_stream_1.default.compose(wrapper1, this.stream, wrapper2);
-        }
-        else if (this.params.mode === "r") {
-            /*  input device  */
-            if (device.maxInputChannels === 0)
-                throw new Error(`device "${device.id}" does not have any input channels (required by read mode)`);
-            this.log("info", `resolved "${this.params.device}" to input device "${device.id}"`);
-            this.io = naudiodon_1.default.AudioIO({
-                inOptions: {
-                    deviceId: device.id,
-                    channelCount: this.config.audioChannels,
-                    sampleRate: this.config.audioSampleRate,
-                    sampleFormat: this.config.audioBitDepth
-                }
-            });
-            this.stream = this.io;
-            /*  convert regular stream into object-mode stream  */
-            const wrapper = utils.createTransformStreamForReadableSide("audio", () => this.timeZero);
-            this.stream.pipe(wrapper);
-            this.stream = wrapper;
-        }
-        else if (this.params.mode === "w") {
-            /*  output device  */
-            if (device.maxOutputChannels === 0)
-                throw new Error(`device "${device.id}" does not have any output channels (required by write mode)`);
-            this.log("info", `resolved "${this.params.device}" to output device "${device.id}"`);
-            this.io = naudiodon_1.default.AudioIO({
-                outOptions: {
-                    deviceId: device.id,
-                    channelCount: this.config.audioChannels,
-                    sampleRate: this.config.audioSampleRate,
-                    sampleFormat: this.config.audioBitDepth
-                }
-            });
-            this.stream = this.io;
-            /*  convert regular stream into object-mode stream  */
-            const wrapper = utils.createTransformStreamForWritableSide();
-            wrapper.pipe(this.stream);
-            this.stream = wrapper;
-        }
-        else
-            throw new Error(`device "${device.id}" does not have any input or output channels`);
-        /*  pass-through PortAudio errors  */
-        this.io.on("error", (err) => {
-            this.emit("error", err);
-        });
-        /*  start PortAudio  */
-        this.io.start();
-    }
-    /*  close node  */
-    async close() {
-        /*  shutdown PortAudio  */
-        if (this.io !== null) {
-            this.io.quit();
-            this.io = null;
-        }
-    }
-}
-exports.default = SpeechFlowNodeDevice;