speechify instead

Author: noahgsolomon

.github/workflows/deploy-ec2.yml

@@ -111,6 +111,8 @@ jobs:
 
           AWS_ACCESS_KEY_ID=${{ secrets.AWS_ACCESS_KEY_ID }}
           AWS_SECRET_ACCESS_KEY=${{ secrets.AWS_SECRET_ACCESS_KEY }}
+
+          SPEECHIFY_API_KEY=${{ secrets.SPEECHIFY_API_KEY }}
           EOF
 
       # Step 7: Build and push the Docker image

generate/audioGenerator.ts

@@ -3,35 +3,38 @@ import fs from 'fs';
 import dotenv from 'dotenv';
 dotenv.config();
 
-const API_BASE_URL = 'https://api.sws.speechify.com';
+const SPEECHIFY_API_URL = 'https://api.sws.speechify.com/v1/audio/speech';
 
 export async function generateAudio(
 	voice_id: string,
 	person: string,
 	line: string,
 	index: number
 ) {
-	const response = await fetch(`${API_BASE_URL}/v1/tts`, {
+	const response = await fetch(`${SPEECHIFY_API_URL}`, {
 		method: 'POST',
 		headers: {
 			'Content-Type': 'application/json',
-			'Authorization': `Bearer ${process.env.SPEECHIFY_API_KEY}`,
+			Authorization: `Bearer ${process.env.SPEECHIFY_API_KEY}`,
 		},
 		body: JSON.stringify({
-			text: line,
+			input: `<speak>${line}</speak>`,
 			voice_id: voice_id,
-			output_format: 'mp3',
-			sample_rate: 24000,
-			speed: 1.0
+			audio_format: 'mp3',
 		}),
 	});
 
 	if (!response.ok) {
 		throw new Error(`Server responded with status code ${response.status}`);
 	}
 
-	// Get the audio buffer from the response
-	const audioBuffer = await response.buffer();
+	const data = await response.json();
+	if (!data.audio_data) {
+		throw new Error('No audio data received from Speechify');
+	}
+
+	// Convert base64 audio data to buffer
+	const audioBuffer = Buffer.from(data.audio_data, 'base64');
 
 	// Write the buffer to a file
 	fs.writeFileSync(`public/voice/${person}-${index}.mp3`, audioBuffer);

generate/speechifyAudioGenerator.ts

@@ -0,0 +1,38 @@
+import fs from 'fs';
+
+const SPEECHIFY_API_URL = 'https://api.sws.speechify.com/v1/audio/speech';
+const SPEECHIFY_API_KEY = 'bcBnuV79sIs3NJZ_TEUI0R3FOxPnu-3YoCBRCodvFKY=';
+
+async function generateSpeechifyAudio(text: string): Promise<Buffer> {
+	console.log('Generating Speechify audio...');
+	const response = await fetch(SPEECHIFY_API_URL, {
+		method: 'POST',
+		headers: {
+			Authorization: `Bearer ${SPEECHIFY_API_KEY}`,
+			'Content-Type': 'application/json',
+		},
+		body: JSON.stringify({
+			input: `<speak>${text}</speak>`,
+			voice_id: '09499a56-f7ba-42f2-996d-c417289c6864',
+			audio_format: 'mp3',
+		}),
+	});
+
+	if (!response.ok) {
+		throw new Error(`Server responded with status code ${response.status}`);
+	}
+
+	const data = await response.json();
+	if (!data.audio_data) {
+		throw new Error('No audio data received from Speechify');
+	}
+
+	return Buffer.from(data.audio_data, 'base64');
+}
+
+async function main() {
+	const audio = await generateSpeechifyAudio('Hello, world!');
+	fs.writeFileSync('output.mp3', new Uint8Array(audio));
+}
+
+main();