whispermix 1.4.0 → 1.4.4

package/README.md

@@ -29,11 +29,13 @@ import WhisperMix from 'whispermix';
 You can initialize WhisperMix with a specific model:
 
 ```javascript
-const whisper = new WhisperMix({ model: 'openai' }); // For OpenAI's Whisper
+const whisper = new WhisperMix({ model: 'openai/whisper-1' }); // For OpenAI's Whisper
 // or
-const whisperGroq = new WhisperMix({ model: 'groq/large-v3' }); // For Groq's Whisper Large v3
+const whisperGroq = new WhisperMix({ model: 'groq/whisper-large-v3' }); // For Groq's Whisper Large v3
 // or
-const whisperLocal = new WhisperMix({ model: 'xenova/large-v3' }); // For local Whisper
+const whisperLocal = new WhisperMix({ model: 'xenova/whisper-large-v3' }); // For local Whisper (large)
+// or
+const whisperLocalBase = new WhisperMix({ model: 'xenova/whisper-base' }); // For local Whisper (base)
 ```
 
 ### 📄 Transcribing from a File
@@ -46,7 +48,7 @@ whisperGroq.fromFile(filePath)
 
 // For local Whisper with language specification
 const whisperLocal = new WhisperMix({ 
-  model: 'xenova/large-v3',
+  model: 'xenova/whisper-large-v3',
   language: 'spanish' // Optional
 });
 whisperLocal.fromFile(filePath)
@@ -79,7 +81,7 @@ WhisperMix uses Bottleneck for rate limiting API-based models. You can configure
 
 ```javascript
 const whisper = new WhisperMix({
-  model: 'whisper-1',
+  model: 'openai/whisper-1',
   bottleneck: {
     minTime: 3000,
     maxConcurrent: 1,
@@ -106,7 +108,7 @@ You can adjust these settings based on your specific rate limiting needs. Note t
 
 Creates a new WhisperMix instance.
 
-- `options.model`: The model to use for transcription. Can be 'openai' (OpenAI), 'groq/large-v3' (Groq), or 'xenova/large-v3' (local).
+- `options.model`: The model to use for transcription. Can be `'openai/whisper-1'` (OpenAI), `'groq/whisper-large-v3'` (Groq), `'xenova/whisper-large-v3'` or `'xenova/whisper-base'` (local).
 - `options.bottleneck`: (Optional) Configuration for Bottleneck rate limiting (API models only).
 - `options.chunkSize`: (Optional) The size in seconds of the chunks to split the audio into. Default is 890 seconds.
 - `options.language`: (Optional) Language for local Whisper model. Defaults to 'auto' for automatic detection.

package/demo/groq.js

@@ -1,17 +1,13 @@
 import 'dotenv/config';
 import WhisperMix from '../index.js';
 
-const transcribe = new WhisperMix({ model: 'whisper-large-v3' });
+const transcribe = new WhisperMix({ model: 'groq/whisper-large-v3' });
 
-async function main() {
-    for (let i = 0; i < 10; i++) {
-        try {
-            const result = await transcribe.fromFile('./example.mp3');
-            console.log(`${i}/10`, result);
-        } catch (error) {
-            console.error(error);
-        }
+for (let i = 0; i < 10; i++) {
+    try {
+        const result = await transcribe.fromFile('./example.mp3');
+        console.log(`${i}/10`, result);
+    } catch (error) {
+        console.error(error);
     }
-}
-
-main();
+}

package/demo/local.js

@@ -1,7 +1,7 @@
 import WhisperMix from '../index.js';
 
 const whisperLocal = new WhisperMix({
-    model: 'xenova/large-v3'
+    model: 'xenova/whisper-large-v3'
 });
 
 const r = await whisperLocal.fromFile('conversation.wav');

package/index.js

@@ -1,5 +1,3 @@
-import axios from 'axios';
-import FormData from 'form-data';
 import fs from 'fs';
 import Bottleneck from 'bottleneck';
 import path from 'path';
@@ -13,7 +11,7 @@ import audioDecode from 'audio-decode';
 
 class WhisperMix {
     constructor(setup = {}) {
-        this.model = 'openai';
+        this.model = 'openai/whisper-1';
         this.bottleneck = {
             minTime: 2000,
             maxConcurrent: 1,
@@ -24,21 +22,21 @@ class WhisperMix {
         this.chunkSize = 15 * 60 - 10; // 14 minutes 50 seconds
         
         const config = {
-            'openai': {
+            'openai/whisper-1': {
                 url: 'https://api.openai.com/v1/audio/transcriptions',
                 modelName: 'whisper-1',
                 apiKey: process.env.OPENAI_API_KEY,
             },
-            'groq/large-v3': {
+            'groq/whisper-large-v3': {
                 url: 'https://api.groq.com/openai/v1/audio/transcriptions',
                 modelName: 'whisper-large-v3',
                 apiKey: process.env.GROQ_API_KEY,
             },
-            'xenova/large-v3': {
+            'xenova/whisper-large-v3': {
                 local: true,
                 modelName: 'Xenova/whisper-large-v3',
             },
-            'xenova/base': {
+            'xenova/whisper-base': {
                 local: true,
                 modelName: 'Xenova/whisper-base',
             },            
@@ -47,6 +45,9 @@ class WhisperMix {
         Object.assign(this, setup);
 
         this.config = config[this.model];
+        if (!this.config) {
+            throw new Error(`Unknown model: "${this.model}". Valid models: ${Object.keys(config).join(', ')}`);
+        }
         this.apiKey = this.apiKey || this.config.apiKey;
         this.apiUrl = this.config.url;
         this.isLocal = this.config.local || false;
@@ -133,29 +134,40 @@ class WhisperMix {
         if (this.isLocal) {
             throw new Error('fromStream is not supported for local Whisper model. Use fromFile instead.');
         }
-        
-        return this.limiter.schedule(() => new Promise((resolve, reject) => {
-            const formData = new FormData();
-            formData.append('file', audioStream);
-            formData.append('model', this.modelName);
 
-            this._makeRequest(formData)
-                .then(resolve)
-                .catch(reject);
-        }));
+        const chunks = [];
+        for await (const chunk of audioStream) {
+            chunks.push(chunk);
+        }
+        const buffer = Buffer.concat(chunks);
+
+        return this.limiter.schedule(() => this._makeRequest(buffer));
     }
 
-    async _makeRequest(formData) {
+    async _makeRequest(buffer) {
         try {
-            const response = await axios.post(this.apiUrl, formData, {
+            const blob = new Blob([buffer], { type: 'audio/mpeg' });
+            const formData = new FormData();
+            formData.append('file', blob, 'audio.mp3');
+            formData.append('model', this.modelName);
+
+            const response = await fetch(this.apiUrl, {
+                method: 'POST',
                 headers: {
-                    ...formData.getHeaders(),
                     'Authorization': `Bearer ${this.apiKey}`,
                 },
+                body: formData,
             });
-            return response.data.text.trim();
+
+            const responseData = await response.json();
+
+            if (!response.ok) {
+                throw responseData;
+            }
+
+            return responseData.text.trim();
         } catch (error) {
-            throw error.response ? error.response.data : error.message;
+            throw error?.message || error;
         }
     }
 

package/package.json

@@ -1,7 +1,7 @@
 {
     "name": "whispermix",
     "description": "🎙️ WhisperMix is a versatile module for transcribing audio using OpenAI’s Whisper or Groq’s Whisper v3 model.",
-    "version": "1.4.0",
+    "version": "1.4.4",
     "type": "module",
     "keywords": [
         "whisper",
@@ -46,10 +46,8 @@
     "dependencies": {
         "@xenova/transformers": "^2.17.2",
         "audio-decode": "^2.1.3",
-        "axios": "^1.12.1",
         "bottleneck": "^2.19.5",
         "fluent-ffmpeg": "^2.1.3",
-        "form-data": "^4.0.4",
         "get-audio-duration": "^4.0.1"
     }
 }