AudioLive
LiveOpenAI Whisper
Accurate multilingual transcription from an audio URL, with plain text, verbose JSON, or timestamped SRT/VTT subtitle output.
Audio → text
$0.012 / minuteLiveOpenAI + Google2 models
Speech to Text API — AI Transcription & Audio Analysis
Muapi's speech-to-text API turns spoken audio into text through one unified REST endpoint. Use Whisper for accurate multilingual transcription with optional timestamped SRT/VTT subtitles, or Gemini Audio Vision for prompted analysis of tone, speakers, and background sound. One API key, one submit-and-poll flow, and pay-as-you-go pricing from $0.012 per minute.
Live audio modelsTranscription and prompted audio understanding2 models
Try Model
AudioLive
LiveGemini Audio Vision
Native audio understanding beyond transcription: ask about speech, tone, speaker changes, music, or background sounds and receive a detailed text analysis.
Audio → analysis
$2.00/M in · $5.00/M outWhat is a Speech to Text API?
A speech-to-text API converts spoken audio into written text through an HTTP request. Submit an audio file URL and receive a transcript without building a manual transcription or audio-processing pipeline.
Muapi exposes Whisper and Gemini Audio Vision through the same unified REST pattern used across the platform. Whisper is the focused choice for multilingual transcription and subtitles; Gemini Audio Vision accepts a prompt when you need analysis, summaries, or structured observations about the whole recording.
Both endpoints use one Muapi API key and the standard asynchronous request-and-poll lifecycle. Choose the model for each job and pay only for the generation you run.
Speech to Text API capabilities
Multilingual transcription
Whisper transcribes spoken audio across dozens of languages and can accept an optional ISO-639-1 language hint.
Timestamped subtitles
Request SRT or VTT output for video captioning, or verbose JSON when you need per-segment timing and metadata.
Prompted audio analysis
Gemini Audio Vision can answer a specific question about an audio file instead of returning only a flat transcript.
Speaker and sound awareness
Ask Gemini to identify speaker changes, tone, emotion, music, and other background sounds in the recording.
Meeting and podcast workflows
Turn interviews, meetings, lectures, and podcasts into searchable transcripts, notes, or structured downstream inputs.
One async integration
Use the same API key, upload flow, submit endpoint, and result polling contract as every other Muapi model.
Speech to text use cases
Video captioning
Generate timestamped SRT or VTT files for videos, courses, podcasts, and social clips.
Meeting intelligence
Transcribe recordings and ask targeted questions about decisions, speakers, action items, or tone.
Voice-driven applications
Feed transcripts into search, agents, summarization, or other text-generation steps in an automated workflow.
Speech to Text Model Comparison
| Model | Provider | Price | Best For |
|---|---|---|---|
| OpenAI Whisper | OpenAI | $0.012 / minute | Accurate multilingual transcription and subtitles |
| Gemini Audio Vision | $2.00/M input · $5.00/M output tokens | Prompted analysis of speech, speakers, tone, and background sound |
Speech to Text API code examples
Submit an audio URL with the model-specific fields, receive a request ID, and poll the standard result endpoint until the transcript or analysis is complete.
1. Transcribe with OpenAI Whisper
curl -X POST https://api.muapi.ai/api/v1/openai-whisper -H "x-api-key: YOUR_API_KEY" -H "Content-Type: application/json" -d '{"audio_url":"https://example.com/meeting.mp3","language":"en","response_format":"verbose_json"}'
# Response: {"request_id":"REQUEST_ID"}
curl https://api.muapi.ai/api/v1/predictions/REQUEST_ID/result -H "x-api-key: YOUR_API_KEY"Pass audio_url and optionally set language, response_format, prompt, or temperature. Use srt or vtt when the output will become subtitles.
2. Analyze audio with Gemini Audio Vision
curl -X POST https://api.muapi.ai/api/v1/gemini-audio-vision -H "x-api-key: YOUR_API_KEY" -H "Content-Type: application/json" -d '{"prompt":"Transcribe this audio and describe speaker changes, tone, and background sounds.","audio_url":"https://example.com/interview.mp3","model":"gemini-2.5-flash"}'
# Response: {"request_id":"REQUEST_ID"}
curl https://api.muapi.ai/api/v1/predictions/REQUEST_ID/result -H "x-api-key: YOUR_API_KEY"Pass audio_url and a prompt describing the transcript, summary, speaker, or sound analysis you need. The result contains generated text.
Speech to Text API FAQ
What is a Speech to Text API?
A speech-to-text API converts an audio recording into written text through an HTTP request. Muapi provides Whisper for accurate transcription and Gemini Audio Vision for prompted audio understanding.
Can Whisper return subtitles instead of plain text?
Yes. Set response_format to srt or vtt for timestamped subtitle output, or use verbose_json for per-segment metadata and timing.
How is Gemini Audio Vision different from transcription?
Gemini Audio Vision accepts a prompt and reasons over the full recording. It can transcribe, summarize, identify speakers, describe tone, and discuss background music or sounds instead of returning only word-for-word text.
How much does speech-to-text cost?
Whisper costs $0.012 per minute of audio with a one-minute minimum. Gemini Audio Vision uses token-based billing at $2.00 per million input tokens and $5.00 per million output tokens.
What audio formats can I submit to Whisper?
Whisper accepts common formats including mp3, mp4, mpeg, mpga, m4a, wav, and webm. The uploaded file must be under 25 MB.
How do I get Speech to Text API access?
Create a Muapi account, generate an API key from the dashboard, and call either endpoint. There is no waitlist.
Text to Speech API →Turn generated text back into natural speech after transcription.AI Audio & Music API →Browse speech, music, sound-effect, and other audio models.AI Lipsync API →Sync a transcript-driven voice track to a face video.
Ready to transcribe audio?
Create one Muapi API key and start with Whisper transcription or Gemini Audio Vision analysis.