Tala
Tala is the Python library and developer toolkit used by the TDM dialogue platform. It provides shared dialogue models, tools for working with dialogue domain descriptions, natural-language generation components, and clients for platform services.
Tala is intended for developers building dialogue applications with TDM or extending components of the platform.
Tala and TDM
TDM (Talkamatic Dialogue Manager) maintains the conversation state and decides how a dialogue proceeds.
A dialogue domain description (DDD) defines the application-specific knowledge and behaviour used by TDM, including questions, actions, dialogue plans, and service interfaces. Tala provides Python representations and supporting tools for these descriptions and for the information exchanged between platform components.
What’s included
- Dialogue models — Python representations of questions, propositions, actions, plans, and other dialogue concepts.
- Domain-description tools — Loading, compilation, and validation of DDDs in supported formats.
- Natural-language generation — Components for producing dialogue utterances.
- Platform clients and utilities — Integration with dialogue services and the Requestor service for language-model requests.
- Interaction testing — Utilities for testing dialogue interactions.
- Interactive command-line client —
tala-interact, for communicating with a running dialogue service.
Installation
Tala currently targets Python 3.12. Install it from PyPI:
python -m pip install tala
Tala provides libraries and a client; installing it does not start a TDM dialogue service. Features that communicate with platform services require access to those services and the corresponding configuration.
Interact with a dialogue service
Use tala-interact to connect to an existing dialogue interaction endpoint:
tala-interact "https://your-host/interact/your-id"
To specify a dialogue domain description:
tala-interact "https://your-host/interact/your-id" --ddd my_ddd
Replace the example URL with the endpoint provided by your dialogue-service deployment. The service must be running and accessible from your machine.
Language-model requests
Tala includes GPTRequest, a Python client for the platform’s Requestor
service, which handles language-model requests. The client supports
OpenAI-compatible response formats, including structured JSON Schema output,
and reports completion status and errors for invalid structured responses.
Using this client requires a configured Requestor endpoint and a model deployment available through that service. See the Requestor client guide for configuration and a complete example.
Documentation and support
For bug reports, include the Tala version, relevant platform component versions, and a minimal example where possible.
Metadata
Release files for tala 31.1.3
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 |
|---|---|---|---|---|
| tala-31.1.3.tar.gz | 364.9 kB | With provenance | Details for tala-31.1.3.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| tala-31.1.3-py3-none-any.whl | Python 3 | none | any | Details for tala-31.1.3-py3-none-any.whl |
Total release size: 609.4 kB
Release files / tala-31.1.3.tar.gz
| Download URL | tala-31.1.3.tar.gz |
|---|---|
| Size | 364.9 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
20d4004e23574c76cb280dc4b88d5adbda680d709672812042490805687f31d9
|
|
BLAKE2b-256 checksum How to use checksums |
ec80e8019e6d5c1e7640b9d0a841215fc8249163d7c03fb7b5f91c6b1df67943
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.16
|
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.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Nov 5, 2026.
Transparency logRelease files / tala-31.1.3-py3-none-any.whl
| Download URL | tala-31.1.3-py3-none-any.whl |
|---|---|
| Size | 244.5 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
b4882fb37b03a43af8a661a7db8c2ac9ec603f8af310d457132468e79b381bf2
|
|
BLAKE2b-256 checksum How to use checksums |
d62e1a176a7bff494e4f277d560b9f499cb64066822d33093d8fa22e93eac466
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.16
|
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.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Nov 5, 2026.
Transparency logSecurity for tala 31.1.3
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.
Full provenance
Both files in this release have been uploaded with provenance metadata.
Attestation for tala-31.1.3.tar.gz
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Nov 5, 2026.
Transparency logA 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.