Model Information

Display Name: HiVox

API Model ID: hinow/hivox

Category: Audio To Text

Description: HiVox turns speech into text. Send an audio file and get the transcript back, billed by the minute of audio. **Key Features:** - Multilingual speech recognition - Accepts a direct file upload or a URL - Redundant upstream routing with automatic failover - Billed by measured audio duration, not by request **Capabilities:** - Transcription of meetings, calls, voice notes and media - Long-form audio in a single request **Best For:** - Turning recorded audio into searchable, indexable text - Feeding transcripts into RAG pipelines **Technical Specs:** - Formats: mp3, wav, flac, ogg, webm, m4a - Max upload: 100 MB - Output: plain text transcript

How to Use This Model

To use HiVox via the HInow.ai API, use the model ID: hinow/hivox

API Request Example (Chat/Text)

POST https://api.hinow.ai/v1/chat/completions
Authorization: Bearer YOUR_API_KEY
Content-Type: application/json

{
  "model": "hinow/hivox",
  "messages": [
    {"role": "user", "content": "Your message here"}
  ]
}
              

Pricing

• per_minute: $0.05

Available Parameters

• audio: URL of the audio file to transcribe. Alternatively, POST multipart/form-data with a 'file' part. Formats: mp3, wav, flac, ogg, webm, m4a. Max 100 MB.
• language: REQUIRED. ISO-639-1 code of the language spoken in the audio, so it is transcribed in that language instead of translated. Regional tags such as 'pt-BR' are accepted and reduced to 'pt'. (Options: pt, en, es, fr, de, it, ja, zh, ko, ru, ar, hi)
• response_format: Output shape. 'json' returns {text}; 'text' returns the bare transcript; 'verbose_json' adds timed segments; 'srt' and 'vtt' return subtitle files. The timed formats (srt, vtt, verbose_json) add $0.02 per minute of audio. (Options: json, text, srt, vtt, verbose_json)

Quick Reference

To use this model, set: "model": "hinow/hivox"

Featured: Yes

Documentation: https://hinow.ai/models/hinow/hivox

API Endpoint: https://api.hinow.ai/v1