Audio and Video Transcript

This Apify actor transcribes audio or video files from publicly accessible URLs using OpenAI's Whisper API. To use this actor, you'll need to provide your own OpenAI API key. It supports multiple languages and highly customizable parameters, enabling precise control over the transcription process. The actor processes each provided URL, downloads the corresponding audio or video files, transcribes them via OpenAI, and securely stores the resulting transcripts in Apify's Storage under the Key-Value Store.

---

🚀 Features

• Automatic language detection or manual language specification from an extensive list.
• Capability to process multiple audio or video URLs simultaneously.
• Versatile output formats including plain text, JSON, SRT, VTT, and verbose JSON.
• Optional inclusion of timestamps for individual words (when using verbose JSON format).
• Fine-tuning through parameters such as temperature, compression ratio thresholds, and speech detection thresholds.
• Secure handling of your OpenAI API key, hidden from logs for added safety.

---

🔧 Input Configuration

Configure your actor with the following parameters:

Parameter	Description	Required
url	Array of publicly accessible audio/video file URLs	✅
language	Language selection or set to Auto-detect	❌
temperature	Floating-point temperature to control variability in transcription	❌
response_format	Desired transcript format (text, srt, vtt, json, verbose_json)	❌
word_timestamps	Include timestamps per word (only valid when using verbose_json format)	❌
prompt	Additional textual context to enhance transcription accuracy	❌
temperature_increment_on_fallback	Increment in temperature if the initial transcription attempt fails	❌
compression_ratio_threshold	Maximum allowable compression ratio for transcript acceptance	❌
logprob_threshold	Minimum log probability required for transcript segments	❌
no_speech_threshold	Probability threshold to detect segments with no speech	❌
openai_api_key	Your personal OpenAI API key (kept secure and hidden)	✅

---

📥 Example Input

1{
2  "url": [
3    { "url": "https://example.com/sample-audio.mp3" }
4  ],
5  "language": "Auto-detect",
6  "temperature": "0.0",
7  "response_format": "text",
8  "word_timestamps": false,
9  "prompt": "",
10  "temperature_increment_on_fallback": 0,
11  "compression_ratio_threshold": 2,
12  "logprob_threshold": -1,
13  "no_speech_threshold": 1,
14  "openai_api_key": "YOUR_OPENAI_API_KEY"
15}

---

📤 Output

Transcription results are securely stored within Apify's Storage under the Key-Value Store. Each transcript is saved individually with an identifiable key for convenient access.