mcpField
T
together-audio

together-audio

github.com/togethercomputerCreative & media3989 installsUpdated 2w ago
Repository

About together-audio

together-audio is a skill for using Together AI audio APIs for speech generation and recognition. It guides developers through REST, streaming, and realtime workflows using supplied references and scripts.

Key features

  1. 01Text-to-speech generation
  2. 02Streaming and realtime voice output
  3. 03Speech transcription and translation
  4. 04Speaker diarization and word timestamps
  5. 05Realtime transcription and captioning

Use cases

  1. 01Generate spoken audio from text
  2. 02Transcribe audio files or URLs
  3. 03Add realtime voice or captions to an app
  4. 04Extract speaker segments and word timings

Install

Copy the command and run it in a terminal.

$npx skills add https://github.com/togethercomputer/skills --skill together-audio

After installing, type/together-audioin Claude Code to confirm it loaded.

Bundled files

References and scripts installed along with SKILL.md.

agents/
openai.yaml
references/
stt-models.md
tts-models.md
scripts/
stt_realtime.py
stt_transcribe.py
stt_transcribe.ts
tts_generate.py
tts_generate.ts
tts_websocket.py

Details

From the project README

Together Audio

Overview

Use Together AI audio APIs for:

  • text-to-speech generation
  • streaming or realtime voice output
  • speech-to-text transcription
  • translation, diarization, and timestamps
  • live captioning and realtime transcription

When This Skill Wins

  • Generate spoken audio from text
  • Transcribe uploaded audio files or URLs
  • Add realtime voice or captioning to an app
  • Extract speaker segments or word timings

Hand Off To Another Skill

  • Use together-chat-completions for text-only generation
  • Use together-video or together-images for visual generation workflows
  • Use together-dedicated-model-inference only when the audio model itself must be hosted on dedicated infrastructure

Quick Routing

Workflow

  1. Confirm whether the task is TTS or STT.
  2. Choose REST, streaming, or realtime transport based on latency and interaction needs.
  3. Pick the model and response format from the relevant reference file.
  4. Start from the matching script instead of rebuilding the request contract from memory.
  5. For Python STT uploads, open audio files in binary mode and pass the file handle rather than a bare path string.

High-Signal Rules

  • Python scripts require the Together v2 SDK (together>=2.0.0). If the user is on an older version, they must upgrade first: uv pip install --upgrade "together>=2.0.0".
  • Use client.audio.speech.create() for TTS.
  • REST TTS returns a BinaryAPIResponse; call response.write_to_file(path) to save it. Do NOT use stream_to_file (it does not exist on this object).
  • Streaming TTS (stream=True) returns a Stream of AudioSpeechStreamChunk objects. Iterate chunks, check chunk.type, and decode base64.b64decode(chunk.delta) for audio data. There is no file-writing helper on the stream object.
  • Use client.audio.transcriptions.create() for transcription and client.audio.translations.create() for translation.
  • Batch transcription and translation share hard limits: 80 MB direct upload, 1 GB URL-fetch, 4 hours of audio per request. For larger payloads, pass a public HTTPS URL on file=; for longer audio, split into ≤ 4 h chunks. See the Limits section of references/stt-models.md.
  • Realtime APIs require audio-format discipline; confirm PCM expectations before streaming bytes.
  • Diarization and word timestamps change response shape; code for the richer verbose output explicitly.

Resource Map

Official Docs

Comments

Markdown supported