"Whisper API Guide 2026: Speech-to-Text for Developers"

from openai import OpenAI

# Using Crazyrouter for competitive pricing on Whisper + 300 other models
client = OpenAI(
    api_key="your-api-key",
    base_url="https://api.crazyrouter.com/v1"
)

# Transcribe an audio file
with open("meeting_recording.mp3", "rb") as audio_file:
    transcript = client.audio.transcriptions.create(
        model="whisper-1",
        file=audio_file,
        response_format="text"
    )

print(transcript)

# Get word-level timestamps
with open("podcast.mp3", "rb") as audio_file:
    transcript = client.audio.transcriptions.create(
        model="whisper-1",
        file=audio_file,
        response_format="verbose_json",
        timestamp_granularities=["word", "segment"]
    )

# Access segments with timestamps
for segment in transcript.segments:
    print(f"[{segment['start']:.1f}s - {segment['end']:.1f}s] {segment['text']}")

# Access word-level timestamps
for word in transcript.words:
    print(f"  {word['word']} ({word['start']:.2f}s)")

# Translate Japanese audio to English text
with open("japanese_interview.mp3", "rb") as audio_file:
    translation = client.audio.translations.create(
        model="whisper-1",
        file=audio_file,
        response_format="text"
    )

print(translation)  # English text output

# Hint the language for improved accuracy
with open("french_lecture.mp3", "rb") as audio_file:
    transcript = client.audio.transcriptions.create(
        model="whisper-1",
        file=audio_file,
        language="fr",  # ISO 639-1 code
        response_format="text"
    )

import OpenAI from 'openai';
import fs from 'fs';

const client = new OpenAI({
    apiKey: 'your-api-key',
    baseURL: 'https://api.crazyrouter.com/v1'
});

// Basic transcription
async function transcribe(filePath) {
    const transcript = await client.audio.transcriptions.create({
        model: 'whisper-1',
        file: fs.createReadStream(filePath),
        response_format: 'verbose_json',
        timestamp_granularities: ['segment']
    });

    transcript.segments.forEach(seg => {
        console.log(`[${seg.start.toFixed(1)}s] ${seg.text}`);
    });

    return transcript;
}

// Translation
async function translateAudio(filePath) {
    const translation = await client.audio.translations.create({
        model: 'whisper-1',
        file: fs.createReadStream(filePath)
    });

    return translation.text;
}

transcribe('meeting.mp3');

# Basic transcription
curl -X POST https://api.crazyrouter.com/v1/audio/transcriptions \
  -H "Authorization: Bearer your-api-key" \
  -H "Content-Type: multipart/form-data" \
  -F file="@recording.mp3" \
  -F model="whisper-1" \
  -F response_format="text"

# With timestamps
curl -X POST https://api.crazyrouter.com/v1/audio/transcriptions \
  -H "Authorization: Bearer your-api-key" \
  -F file="@recording.mp3" \
  -F model="whisper-1" \
  -F response_format="verbose_json" \
  -F 'timestamp_granularities[]=word'

from pydub import AudioSegment

def split_audio(file_path, chunk_length_ms=600000):  # 10-minute chunks
    audio = AudioSegment.from_file(file_path)
    chunks = []
    for i in range(0, len(audio), chunk_length_ms):
        chunk = audio[i:i + chunk_length_ms]
        chunk_path = f"chunk_{i // chunk_length_ms}.mp3"
        chunk.export(chunk_path, format="mp3")
        chunks.append(chunk_path)
    return chunks

# Transcribe all chunks
chunks = split_audio("long_recording.mp3")
full_transcript = ""
for chunk_path in chunks:
    with open(chunk_path, "rb") as f:
        result = client.audio.transcriptions.create(
            model="whisper-1", file=f, response_format="text"
        )
        full_transcript += result + " "

# Use the prompt parameter to guide recognition of specific terms
with open("tech_meeting.mp3", "rb") as audio_file:
    transcript = client.audio.transcriptions.create(
        model="whisper-1",
        file=audio_file,
        prompt="Crazyrouter, GPT-5, LangChain, Kubernetes, PostgreSQL, NGINX",
        response_format="text"
    )