This release is a pre-release and may not be stable for production use.
swarmauri_tts_together
Standalone alpha together speech synthesis for Swarmauri. Together /v1/audio/speech: Orpheus and Kokoro. HTTP file synthesis and SSE audio streaming are implemented. WebSocket sessions are excluded.
Features
- File generation through
predictandapredict. - Incremental audio through
streamandastream, with connection cleanup. - Synchronous batch execution and bounded async batches.
- Explicit provider URL, model, voice and output settings.
- No automatic retries of potentially billable generation requests.
Together streaming always returns decoded raw PCM (16-bit little-endian at sample_rate, default 24000 Hz), while response_format controls complete file output. SSE framing is parsed rather than returned as audio.
Installation
uv add --prerelease allow swarmauri_tts_together
pip install --pre swarmauri_tts_together
This source candidate requires the companion swarmauri_base release providing
HttpTTSMixin (>=0.11.0.dev8). Release that dependency before these wheels.
Usage
Set TOGETHER_API_KEY in your environment, then:
import os
from swarmauri_tts_together import TogetherTTS
provider = TogetherTTS(
api_key=os.environ["TOGETHER_API_KEY"],
name="canopylabs/orpheus-3b-0.1-ft",
voice="tara",
)
path = provider.predict("Welcome to Swarmauri.", "welcome.mp3")
# await provider.apredict(text, audio_path)
# provider.batch({"First message": "first.mp3"})
# await provider.abatch({"Second message": "second.mp3"}, max_concurrent=2)
# Iterate provider.stream(text), or async-iterate provider.astream(text).
Choose voices appropriate to the selected model. The filename does not change the
codec: select response_format explicitly and use a matching extension. The
component implements TTSBase and can be passed to TextToSpeechAgent.
Verification and alpha status
Live API behavior has not been verified by this change.
Tracked in verification issue #6270. Offline tests validate
payloads, errors and transport behavior; they do not establish speech quality,
model availability or actual billed cost. Keep Development Status :: 3 - Alpha
in pyproject.toml until maintainers review live verification evidence.
Live tests are disabled even if a key is present. To explicitly authorize one
billable synthesis request, set SWARMAURI_LIVE_TTS=1, set TOGETHER_API_KEY privately,
and run this package's tests/i9n suite. Only predict runs by default. Set
SWARMAURI_LIVE_TTS_OPERATIONS=predict,apredict,stream,astream,batch,abatch to
exercise all operations (eight requests including the two-item batches). Set
SWARMAURI_TTS_MODEL and SWARMAURI_TTS_VOICE to select another model/voice.
Do not paste API keys into issues or logs.
Verification issue links and evidence requirements are maintained in the verification guide.
License
Apache-2.0. Part of the Swarmauri SDK.
Metadata
Release files for swarmauri_tts_together 0.1.0a2
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| swarmauri_tts_together-0.1.0a2.tar.gz | 8.3 kB | Without provenance | Details for swarmauri_tts_together-0.1.0a2.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| swarmauri_tts_together-0.1.0a2-py3-none-any.whl | Python 3 | none | any | Details for swarmauri_tts_together-0.1.0a2-py3-none-any.whl |
Total release size: 18.3 kB
Release files / swarmauri_tts_together-0.1.0a2.tar.gz
| Download URL | swarmauri_tts_together-0.1.0a2.tar.gz |
|---|---|
| Size | 8.3 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
506ba8419493a7efd09100ce45841a65716f97807b544f134a1bbe1f955ebe10
|
|
BLAKE2b-256 checksum How to use checksums |
fa315dd0f44513a30a9edd86a8a87029215b2ef61cf82766839b7530f3e74184
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
uv/0.13.9 {"installer":{"name":"uv","version":"0.13.9","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"Ubuntu","version":"24.04","id":"noble","libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":true}
|
Release files / swarmauri_tts_together-0.1.0a2-py3-none-any.whl
| Download URL | swarmauri_tts_together-0.1.0a2-py3-none-any.whl |
|---|---|
| Size | 10.0 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
301af042f3949dc08b7939332938e038fcba5b0a8c852c6434d5b25da316428f
|
|
BLAKE2b-256 checksum How to use checksums |
29544c0e3f8e587d419454c954aaf13a53d1f717a39271828f1e82d498e5b8f8
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
uv/0.13.9 {"installer":{"name":"uv","version":"0.13.9","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"Ubuntu","version":"24.04","id":"noble","libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":true}
|
Security for swarmauri_tts_together 0.1.0a2
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.