Skip to main content

soniox-transcribe

CLI tool for generating subtitles from audio/video files using the Soniox Speech-to-Text API.

Features

  • Subtitle formats: SRT, VTT, TXT
  • Video support: Automatically extracts audio from video files via ffmpeg
  • Speaker diarization: Speaker labels enabled by default
  • 60+ languages: With automatic language identification
  • Context support: Custom terms and background text to improve accuracy (up to ~8,000 tokens)

Installation

pip install soniox-transcribe

For video file support, ffmpeg must be installed:

# macOS
brew install ffmpeg

# Ubuntu/Debian
sudo apt install ffmpeg

Configuration

Set your Soniox API key as an environment variable:

export SONIOX_API_KEY=<your-api-key>

Get your API key at console.soniox.com.

Alternatively, create a .env file in your working directory:

SONIOX_API_KEY=<your-api-key>

Usage

# Basic transcription (outputs .srt by default)
soniox-transcribe video.mp4

# Specify output file (format inferred from extension)
soniox-transcribe video.mp4 -o video.vtt

# Explicit format
soniox-transcribe audio.wav -f txt

# Transcribe from URL
soniox-transcribe --url https://example.com/audio.mp3

# Disable speaker diarization
soniox-transcribe interview.wav --no-speaker

# Add custom vocabulary
soniox-transcribe lecture.mp3 -t "Transformer" -t "BERT" -t "GPT"

# Provide background text for better accuracy
soniox-transcribe meeting.mp3 --context-text "Discussion about Q4 revenue targets"

# Use a context file
soniox-transcribe meeting.mp3 --context-text-file agenda.txt

Options

Option Description
-o, --output Output file path (default: input with .srt extension)
-f, --format Output format: srt, vtt, txt
-u, --url Transcribe from a public URL instead of a local file
-l, --language Language hints (default: zh, en); repeatable
--no-speaker Disable speaker diarization and labels
--max-chars Max characters per subtitle line (default: 42)
-m, --model Soniox model (default: stt-async-v4)
-t, --context-term Custom terms/vocabulary; repeatable
--context-text Background text for accuracy improvement
--context-text-file File containing background text
-c, --context JSON string with full context configuration
--context-file JSON file with full context configuration

Supported Formats

Audio: mp3, wav, flac, aac, ogg, m4a, wma

Video: mp4, mkv, avi, mov, webm, flv, wmv, m4v (requires ffmpeg)

License

MIT

Metadata

Release files for soniox-transcribe 0.1.0

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for soniox-transcribe 0.1.0
File Size Uploaded Provenance Actions
soniox_transcribe-0.1.0.tar.gz 10.2 kB Without provenance Details for soniox_transcribe-0.1.0.tar.gz

Built distribution (wheel)

Table of built distributions (wheels) for soniox-transcribe 0.1.0
File Interpreter ABI Platform Actions
soniox_transcribe-0.1.0-py3-none-any.whl Python 3 none any Details for soniox_transcribe-0.1.0-py3-none-any.whl

Total release size: 23.3 kB

Release files / soniox_transcribe-0.1.0.tar.gz

Download URL soniox_transcribe-0.1.0.tar.gz
Size 10.2 kB
Tags Source
SHA-256 checksum
How to use checksums
624f67c1830ab98180e526f91752159300dadcad27a5cef81a741ecd6c7347a9
BLAKE2b-256 checksum
How to use checksums
f98c8734173658e38207f7ca81315253d3cf781d1a5a5463f439bc6f7ea50b0d
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/6.2.0 CPython/3.12.9

Release files / soniox_transcribe-0.1.0-py3-none-any.whl

Download URL soniox_transcribe-0.1.0-py3-none-any.whl
Size 13.1 kB
Tags Python 3
SHA-256 checksum
How to use checksums
1450b47cc2e689c362c9514381d4811d22a96d5bb65ceaaa655a665cca6619f6
BLAKE2b-256 checksum
How to use checksums
7db63648dfff1a64f0da6eea79632459964fb4a2b786815e8b7c37d5e95744fa
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/6.2.0 CPython/3.12.9

Security for soniox-transcribe 0.1.0

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

No provenance found

None of the files in this release were uploaded with provenance metadata.

A note on package security

PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.

To learn more about protecting yourself from malicious actors, visit our package security documentation.

Release history Release notifications | RSS feed

This release

0.1.0 This release

2 release files No provenance

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page