feat: finishing touches TTS live client

Author: SandraRodgers

examples/node-speak-live/index.js

@@ -1,44 +1,62 @@
+const fs = require("fs");
 const { createClient, LiveTTSEvents } = require("../../dist/main/index");
-const fetch = require("cross-fetch");
 
 const live = async () => {
   const text = "Hello, how can I help you today?";
 
-  const deepgram = createClient(process.env.DEEPGRAM_API_KEY, {
-    global: { fetch: { options: { url: "https://api.beta.deepgram.com" } } },
-  });
+  const deepgram = createClient(process.env.DEEPGRAM_API_KEY);
+
+  const connection = deepgram.speak.live({ model: "aura-asteria-en" });
 
-  const connection = deepgram.speak.live({ text }, { model: "aura-asteria-en" });
+  let audioBuffer = Buffer.alloc(0);
 
   connection.on(LiveTTSEvents.Open, () => {
+    console.log("Connection opened");
+
+    // Send text data for TTS synthesis
+    connection.sendText(text);
+
+    // Send Flush message to the server after sending the text
+    connection.flush();
+
     connection.on(LiveTTSEvents.Close, () => {
-      console.log("Connection closed.");
+      console.log("Connection closed");
     });
 
     connection.on(LiveTTSEvents.Metadata, (data) => {
-      console.log(`Deepgram Metadata: ${data}`);
+      console.dir(data, { depth: null });
     });
 
     connection.on(LiveTTSEvents.Audio, (data) => {
-      console.log(`Deepgram Audio: ${data}`);
+      console.log("Deepgram audio data received");
+      // Concatenate the audio chunks into a single buffer
+      const buffer = Buffer.from(data);
+      audioBuffer = Buffer.concat([audioBuffer, buffer]);
     });
 
-    connection.on(LiveTTSEvents.Flushed, (data) => {
+    connection.on(LiveTTSEvents.Flushed, () => {
       console.log("Deepgram Flushed");
+      // Write the buffered audio data to a file when the flush event is received
+      writeFile();
     });
 
     connection.on(LiveTTSEvents.Error, (err) => {
       console.error(err);
     });
+  });
 
-    fetch(url)
-      .then((r) => r.body)
-      .then((res) => {
-        res.on("readable", () => {
-          connection.send(res.read());
-        });
+  const writeFile = () => {
+    if (audioBuffer.length > 0) {
+      fs.writeFile("output.mp3", audioBuffer, (err) => {
+        if (err) {
+          console.error("Error writing audio file:", err);
+        } else {
+          console.log("Audio file saved as output.mp3");
+        }
       });
-  });
+      audioBuffer = Buffer.alloc(0); // Reset buffer after writing
+    }
+  };
 };
 
 live();

src/packages/AbstractLiveClient.ts

@@ -3,7 +3,6 @@ import { CONNECTION_STATE, SOCKET_STATES } from "../lib/constants";
 import type { DeepgramClientOptions, LiveSchema } from "../lib/types";
 import type { WebSocket as WSWebSocket } from "ws";
 import { isBun } from "../lib/helpers";
-import { LiveTTSEvents } from "../lib/enums";
 
 /**
  * Represents a constructor for a WebSocket-like object that can be used in the application.
@@ -277,48 +276,6 @@ export abstract class AbstractLiveClient extends AbstractClient {
    * @abstract Requires subclasses to set up context aware event handlers.
    */
   abstract setupConnection(): void;
-
-  /**
-   * Handles incoming messages from the WebSocket connection.
-   * @param event - The MessageEvent object representing the received message.
-   */
-  protected handleMessage(event: MessageEvent): void {
-    if (typeof event.data === "string") {
-      try {
-        const data = JSON.parse(event.data);
-        this.handleTextMessage(data);
-      } catch (error) {
-        this.emit(LiveTTSEvents.Error, {
-          event,
-          message: "Unable to parse `data` as JSON.",
-          error,
-        });
-      }
-    } else if (event.data instanceof ArrayBuffer) {
-      this.handleBinaryMessage(event.data);
-    } else {
-      this.emit(LiveTTSEvents.Error, {
-        event,
-        message: "Received unknown data type.",
-      });
-    }
-  }
-
-  /**
-   * Handles text messages received from the WebSocket connection.
-   * @param data - The parsed JSON data.
-   */
-  protected handleTextMessage(data: any): void {
-    // To be implemented by subclasses
-  }
-
-  /**
-   * Handles binary messages received from the WebSocket connection.
-   * @param data - The binary data.
-   */
-  protected handleBinaryMessage(data: ArrayBuffer): void {
-    // To be implemented by subclasses
-  }
 }
 
 class WSWebSocketDummy {

src/packages/SpeakClient.ts

@@ -2,6 +2,7 @@ import { AbstractClient } from "./AbstractClient";
 import { SpeakLiveClient } from "./SpeakLiveClient";
 import { SpeakRestClient } from "./SpeakRestClient";
 import { SpeakSchema } from "../lib/types";
+import { TextSource } from "../lib/types";
 
 /**
  * The `SpeakClient` class extends the `AbstractClient` class and provides access to the "speak" namespace.
@@ -16,8 +17,10 @@ export class SpeakClient extends AbstractClient {
   /**
    * Returns a `SpeakRestClient` instance for interacting with the rest speak API.
    */
-  get request() {
-    return new SpeakRestClient(this.options);
+  public request(source: TextSource, options?: SpeakSchema, endpoint = ":version/speak") {
+    const client = new SpeakRestClient(this.options);
+
+    return client.request(source, options, endpoint);
   }
 
   /**

src/packages/SpeakLiveClient.ts

@@ -11,9 +11,7 @@ import type { SpeakSchema, DeepgramClientOptions } from "../lib/types";
  *
  * The `configure` method allows you to send additional configuration options to the connected session.
  *
- *
  * The `requestClose` method requests the server to close the connection.
- *
  */
 export class SpeakLiveClient extends AbstractLiveClient {
   public namespace: string = "speak";
@@ -133,4 +131,33 @@ export class SpeakLiveClient extends AbstractLiveClient {
       })
     );
   }
+
+  /**
+   * Handles incoming messages from the WebSocket connection.
+   * @param event - The MessageEvent object representing the received message.
+   */
+  protected handleMessage(event: MessageEvent): void {
+    if (typeof event.data === "string") {
+      try {
+        const data = JSON.parse(event.data);
+        this.handleTextMessage(data);
+      } catch (error) {
+        this.emit(LiveTTSEvents.Error, {
+          event,
+          message: "Unable to parse `data` as JSON.",
+          error,
+        });
+      }
+    } else if (event.data instanceof ArrayBuffer) {
+      this.handleBinaryMessage(event.data);
+    } else if (Buffer.isBuffer(event.data)) {
+      this.handleBinaryMessage(event.data.buffer);
+    } else {
+      console.log("Received unknown data type", event.data);
+      this.emit(LiveTTSEvents.Error, {
+        event,
+        message: "Received unknown data type.",
+      });
+    }
+  }
 }

src/packages/index.ts

@@ -9,3 +9,5 @@ export * from "./ModelsRestClient";
 export * from "./ReadRestClient";
 export * from "./SelfHostedRestClient";
 export * from "./SpeakClient";
+export * from "./SpeakLiveClient";
+export * from "./SpeakRestClient";