wyoming-mlx 0.1.0__tar.gz

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (42) hide show
  1. wyoming_mlx-0.1.0/.github/workflows/ci.yml +30 -0
  2. wyoming_mlx-0.1.0/.github/workflows/release.yml +88 -0
  3. wyoming_mlx-0.1.0/.gitignore +13 -0
  4. wyoming_mlx-0.1.0/.python-version +1 -0
  5. wyoming_mlx-0.1.0/LICENSE +202 -0
  6. wyoming_mlx-0.1.0/PKG-INFO +170 -0
  7. wyoming_mlx-0.1.0/README.md +140 -0
  8. wyoming_mlx-0.1.0/mise.toml +3 -0
  9. wyoming_mlx-0.1.0/pyproject.toml +102 -0
  10. wyoming_mlx-0.1.0/scripts/dev_run.py +28 -0
  11. wyoming_mlx-0.1.0/scripts/smoke.py +86 -0
  12. wyoming_mlx-0.1.0/src/wyoming_mlx/__init__.py +0 -0
  13. wyoming_mlx-0.1.0/src/wyoming_mlx/__main__.py +319 -0
  14. wyoming_mlx-0.1.0/src/wyoming_mlx/auth.py +38 -0
  15. wyoming_mlx-0.1.0/src/wyoming_mlx/backends/__init__.py +0 -0
  16. wyoming_mlx-0.1.0/src/wyoming_mlx/backends/base.py +31 -0
  17. wyoming_mlx-0.1.0/src/wyoming_mlx/backends/fake.py +41 -0
  18. wyoming_mlx-0.1.0/src/wyoming_mlx/backends/mlx_stt.py +63 -0
  19. wyoming_mlx-0.1.0/src/wyoming_mlx/backends/mlx_tts.py +137 -0
  20. wyoming_mlx-0.1.0/src/wyoming_mlx/config.py +100 -0
  21. wyoming_mlx-0.1.0/src/wyoming_mlx/http/__init__.py +1 -0
  22. wyoming_mlx-0.1.0/src/wyoming_mlx/http/app.py +25 -0
  23. wyoming_mlx-0.1.0/src/wyoming_mlx/http/routes.py +153 -0
  24. wyoming_mlx-0.1.0/src/wyoming_mlx/wyoming_servers/__init__.py +0 -0
  25. wyoming_mlx-0.1.0/src/wyoming_mlx/wyoming_servers/stt.py +79 -0
  26. wyoming_mlx-0.1.0/src/wyoming_mlx/wyoming_servers/tts.py +67 -0
  27. wyoming_mlx-0.1.0/tests/__init__.py +0 -0
  28. wyoming_mlx-0.1.0/tests/conftest.py +25 -0
  29. wyoming_mlx-0.1.0/tests/integration/test_roundtrip.py +40 -0
  30. wyoming_mlx-0.1.0/tests/unit/__init__.py +0 -0
  31. wyoming_mlx-0.1.0/tests/unit/backends/__init__.py +0 -0
  32. wyoming_mlx-0.1.0/tests/unit/backends/test_fake.py +37 -0
  33. wyoming_mlx-0.1.0/tests/unit/backends/test_mlx_stt.py +40 -0
  34. wyoming_mlx-0.1.0/tests/unit/http/__init__.py +0 -0
  35. wyoming_mlx-0.1.0/tests/unit/http/test_routes.py +201 -0
  36. wyoming_mlx-0.1.0/tests/unit/test_auth.py +51 -0
  37. wyoming_mlx-0.1.0/tests/unit/test_config.py +51 -0
  38. wyoming_mlx-0.1.0/tests/unit/test_main.py +26 -0
  39. wyoming_mlx-0.1.0/tests/unit/wyoming_servers/__init__.py +0 -0
  40. wyoming_mlx-0.1.0/tests/unit/wyoming_servers/test_stt.py +148 -0
  41. wyoming_mlx-0.1.0/tests/unit/wyoming_servers/test_tts.py +116 -0
  42. wyoming_mlx-0.1.0/uv.lock +2644 -0
@@ -0,0 +1,30 @@
1
+ name: CI
2
+
3
+ on:
4
+ push:
5
+ branches: [main]
6
+ pull_request:
7
+
8
+ jobs:
9
+ test:
10
+ runs-on: ubuntu-latest
11
+
12
+ steps:
13
+ - uses: actions/checkout@v5
14
+
15
+ - uses: jdx/mise-action@v2
16
+
17
+ - name: Install dependencies
18
+ run: uv sync
19
+
20
+ - name: Lint
21
+ run: uv run ruff check .
22
+
23
+ - name: Format check
24
+ run: uv run ruff format --check .
25
+
26
+ - name: Type check
27
+ run: uv run pyright
28
+
29
+ - name: Test with coverage
30
+ run: uv run pytest --cov --cov-report=term-missing
@@ -0,0 +1,88 @@
1
+ name: Release
2
+
3
+ on:
4
+ release:
5
+ types: [published]
6
+ workflow_dispatch:
7
+ inputs:
8
+ tag:
9
+ description: "Tag to release (e.g. v0.1.0)"
10
+ required: true
11
+ publish_pypi:
12
+ description: "Also publish to PyPI"
13
+ type: boolean
14
+ default: false
15
+
16
+ jobs:
17
+ pypi:
18
+ if: github.event_name == 'release' || inputs.publish_pypi
19
+ runs-on: ubuntu-latest
20
+ environment: pypi
21
+ permissions:
22
+ id-token: write
23
+ contents: read
24
+ steps:
25
+ - uses: actions/checkout@v5
26
+ with:
27
+ ref: ${{ github.event.release.tag_name || inputs.tag }}
28
+
29
+ - uses: astral-sh/setup-uv@v8.2.0
30
+
31
+ - name: Build sdist and wheel
32
+ run: uv build
33
+
34
+ - name: Publish to PyPI (trusted publishing)
35
+ run: uv publish --trusted-publishing always
36
+
37
+ homebrew:
38
+ runs-on: ubuntu-latest
39
+ permissions:
40
+ contents: read
41
+ steps:
42
+ - name: Check out tap
43
+ uses: actions/checkout@v5
44
+ with:
45
+ repository: rnorth/homebrew-tap
46
+ ssh-key: ${{ secrets.TAP_DEPLOY_KEY }}
47
+
48
+ - name: Compute release tarball sha256
49
+ id: tarball
50
+ env:
51
+ GH_TOKEN: ${{ github.token }}
52
+ TAG: ${{ github.event.release.tag_name || inputs.tag }}
53
+ run: |
54
+ url="https://github.com/${{ github.repository }}/archive/refs/tags/${TAG}.tar.gz"
55
+ out="$RUNNER_TEMP/release.tar.gz"
56
+ # Unauthenticated first (public repo), then token auth (private repo).
57
+ code=$(curl -sL -o "$out" -w "%{http_code}" "$url")
58
+ if [ "$code" != "200" ]; then
59
+ echo "Unauthenticated fetch: HTTP $code; retrying with token"
60
+ code=$(curl -sL -H "Authorization: Bearer $GH_TOKEN" -o "$out" -w "%{http_code}" "$url")
61
+ fi
62
+ if [ "$code" != "200" ]; then
63
+ echo "::error::Could not fetch $url (HTTP $code)"
64
+ exit 1
65
+ fi
66
+ echo "sha=$(sha256sum "$out" | cut -d' ' -f1)" >> "$GITHUB_OUTPUT"
67
+
68
+ - name: Update formula
69
+ env:
70
+ TAG: ${{ github.event.release.tag_name || inputs.tag }}
71
+ SHA: ${{ steps.tarball.outputs.sha }}
72
+ run: |
73
+ sed -i -E "s|/archive/refs/tags/v[0-9][^\"]*\.tar\.gz|/archive/refs/tags/${TAG}.tar.gz|" Formula/wyoming-mlx.rb
74
+ sed -i -E "s|sha256 \"[0-9a-f]{64}\"|sha256 \"${SHA}\"|" Formula/wyoming-mlx.rb
75
+ git diff
76
+
77
+ - name: Commit and push
78
+ env:
79
+ TAG: ${{ github.event.release.tag_name || inputs.tag }}
80
+ run: |
81
+ if git diff --quiet; then
82
+ echo "Formula already up to date."
83
+ exit 0
84
+ fi
85
+ git config user.name "github-actions[bot]"
86
+ git config user.email "github-actions[bot]@users.noreply.github.com"
87
+ git commit -am "wyoming-mlx ${TAG}"
88
+ git push
@@ -0,0 +1,13 @@
1
+ __pycache__/
2
+ *.py[cod]
3
+ .venv/
4
+ .pytest_cache/
5
+ .ruff_cache/
6
+ .pyright/
7
+ dist/
8
+ build/
9
+ *.egg-info/
10
+ models/
11
+ notes.md
12
+ _dev_run.py
13
+ .coverage
@@ -0,0 +1 @@
1
+ 3.12
@@ -0,0 +1,202 @@
1
+
2
+ Apache License
3
+ Version 2.0, January 2004
4
+ http://www.apache.org/licenses/
5
+
6
+ TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
7
+
8
+ 1. Definitions.
9
+
10
+ "License" shall mean the terms and conditions for use, reproduction,
11
+ and distribution as defined by Sections 1 through 9 of this document.
12
+
13
+ "Licensor" shall mean the copyright owner or entity authorized by
14
+ the copyright owner that is granting the License.
15
+
16
+ "Legal Entity" shall mean the union of the acting entity and all
17
+ other entities that control, are controlled by, or are under common
18
+ control with that entity. For the purposes of this definition,
19
+ "control" means (i) the power, direct or indirect, to cause the
20
+ direction or management of such entity, whether by contract or
21
+ otherwise, or (ii) ownership of fifty percent (50%) or more of the
22
+ outstanding shares, or (iii) beneficial ownership of such entity.
23
+
24
+ "You" (or "Your") shall mean an individual or Legal Entity
25
+ exercising permissions granted by this License.
26
+
27
+ "Source" form shall mean the preferred form for making modifications,
28
+ including but not limited to software source code, documentation
29
+ source, and configuration files.
30
+
31
+ "Object" form shall mean any form resulting from mechanical
32
+ transformation or translation of a Source form, including but
33
+ not limited to compiled object code, generated documentation,
34
+ and conversions to other media types.
35
+
36
+ "Work" shall mean the work of authorship, whether in Source or
37
+ Object form, made available under the License, as indicated by a
38
+ copyright notice that is included in or attached to the work
39
+ (an example is provided in the Appendix below).
40
+
41
+ "Derivative Works" shall mean any work, whether in Source or Object
42
+ form, that is based on (or derived from) the Work and for which the
43
+ editorial revisions, annotations, elaborations, or other modifications
44
+ represent, as a whole, an original work of authorship. For the purposes
45
+ of this License, Derivative Works shall not include works that remain
46
+ separable from, or merely link (or bind by name) to the interfaces of,
47
+ the Work and Derivative Works thereof.
48
+
49
+ "Contribution" shall mean any work of authorship, including
50
+ the original version of the Work and any modifications or additions
51
+ to that Work or Derivative Works thereof, that is intentionally
52
+ submitted to Licensor for inclusion in the Work by the copyright owner
53
+ or by an individual or Legal Entity authorized to submit on behalf of
54
+ the copyright owner. For the purposes of this definition, "submitted"
55
+ means any form of electronic, verbal, or written communication sent
56
+ to the Licensor or its representatives, including but not limited to
57
+ communication on electronic mailing lists, source code control systems,
58
+ and issue tracking systems that are managed by, or on behalf of, the
59
+ Licensor for the purpose of discussing and improving the Work, but
60
+ excluding communication that is conspicuously marked or otherwise
61
+ designated in writing by the copyright owner as "Not a Contribution."
62
+
63
+ "Contributor" shall mean Licensor and any individual or Legal Entity
64
+ on behalf of whom a Contribution has been received by Licensor and
65
+ subsequently incorporated within the Work.
66
+
67
+ 2. Grant of Copyright License. Subject to the terms and conditions of
68
+ this License, each Contributor hereby grants to You a perpetual,
69
+ worldwide, non-exclusive, no-charge, royalty-free, irrevocable
70
+ copyright license to reproduce, prepare Derivative Works of,
71
+ publicly display, publicly perform, sublicense, and distribute the
72
+ Work and such Derivative Works in Source or Object form.
73
+
74
+ 3. Grant of Patent License. Subject to the terms and conditions of
75
+ this License, each Contributor hereby grants to You a perpetual,
76
+ worldwide, non-exclusive, no-charge, royalty-free, irrevocable
77
+ (except as stated in this section) patent license to make, have made,
78
+ use, offer to sell, sell, import, and otherwise transfer the Work,
79
+ where such license applies only to those patent claims licensable
80
+ by such Contributor that are necessarily infringed by their
81
+ Contribution(s) alone or by combination of their Contribution(s)
82
+ with the Work to which such Contribution(s) was submitted. If You
83
+ institute patent litigation against any entity (including a
84
+ cross-claim or counterclaim in a lawsuit) alleging that the Work
85
+ or a Contribution incorporated within the Work constitutes direct
86
+ or contributory patent infringement, then any patent licenses
87
+ granted to You under this License for that Work shall terminate
88
+ as of the date such litigation is filed.
89
+
90
+ 4. Redistribution. You may reproduce and distribute copies of the
91
+ Work or Derivative Works thereof in any medium, with or without
92
+ modifications, and in Source or Object form, provided that You
93
+ meet the following conditions:
94
+
95
+ (a) You must give any other recipients of the Work or
96
+ Derivative Works a copy of this License; and
97
+
98
+ (b) You must cause any modified files to carry prominent notices
99
+ stating that You changed the files; and
100
+
101
+ (c) You must retain, in the Source form of any Derivative Works
102
+ that You distribute, all copyright, patent, trademark, and
103
+ attribution notices from the Source form of the Work,
104
+ excluding those notices that do not pertain to any part of
105
+ the Derivative Works; and
106
+
107
+ (d) If the Work includes a "NOTICE" text file as part of its
108
+ distribution, then any Derivative Works that You distribute must
109
+ include a readable copy of the attribution notices contained
110
+ within such NOTICE file, excluding those notices that do not
111
+ pertain to any part of the Derivative Works, in at least one
112
+ of the following places: within a NOTICE text file distributed
113
+ as part of the Derivative Works; within the Source form or
114
+ documentation, if provided along with the Derivative Works; or,
115
+ within a display generated by the Derivative Works, if and
116
+ wherever such third-party notices normally appear. The contents
117
+ of the NOTICE file are for informational purposes only and
118
+ do not modify the License. You may add Your own attribution
119
+ notices within Derivative Works that You distribute, alongside
120
+ or as an addendum to the NOTICE text from the Work, provided
121
+ that such additional attribution notices cannot be construed
122
+ as modifying the License.
123
+
124
+ You may add Your own copyright statement to Your modifications and
125
+ may provide additional or different license terms and conditions
126
+ for use, reproduction, or distribution of Your modifications, or
127
+ for any such Derivative Works as a whole, provided Your use,
128
+ reproduction, and distribution of the Work otherwise complies with
129
+ the conditions stated in this License.
130
+
131
+ 5. Submission of Contributions. Unless You explicitly state otherwise,
132
+ any Contribution intentionally submitted for inclusion in the Work
133
+ by You to the Licensor shall be under the terms and conditions of
134
+ this License, without any additional terms or conditions.
135
+ Notwithstanding the above, nothing herein shall supersede or modify
136
+ the terms of any separate license agreement you may have executed
137
+ with Licensor regarding such Contributions.
138
+
139
+ 6. Trademarks. This License does not grant permission to use the trade
140
+ names, trademarks, service marks, or product names of the Licensor,
141
+ except as required for reasonable and customary use in describing the
142
+ origin of the Work and reproducing the content of the NOTICE file.
143
+
144
+ 7. Disclaimer of Warranty. Unless required by applicable law or
145
+ agreed to in writing, Licensor provides the Work (and each
146
+ Contributor provides its Contributions) on an "AS IS" BASIS,
147
+ WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
148
+ implied, including, without limitation, any warranties or conditions
149
+ of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
150
+ PARTICULAR PURPOSE. You are solely responsible for determining the
151
+ appropriateness of using or redistributing the Work and assume any
152
+ risks associated with Your exercise of permissions under this License.
153
+
154
+ 8. Limitation of Liability. In no event and under no legal theory,
155
+ whether in tort (including negligence), contract, or otherwise,
156
+ unless required by applicable law (such as deliberate and grossly
157
+ negligent acts) or agreed to in writing, shall any Contributor be
158
+ liable to You for damages, including any direct, indirect, special,
159
+ incidental, or consequential damages of any character arising as a
160
+ result of this License or out of the use or inability to use the
161
+ Work (including but not limited to damages for loss of goodwill,
162
+ work stoppage, computer failure or malfunction, or any and all
163
+ other commercial damages or losses), even if such Contributor
164
+ has been advised of the possibility of such damages.
165
+
166
+ 9. Accepting Warranty or Additional Liability. While redistributing
167
+ the Work or Derivative Works thereof, You may choose to offer,
168
+ and charge a fee for, acceptance of support, warranty, indemnity,
169
+ or other liability obligations and/or rights consistent with this
170
+ License. However, in accepting such obligations, You may act only
171
+ on Your own behalf and on Your sole responsibility, not on behalf
172
+ of any other Contributor, and only if You agree to indemnify,
173
+ defend, and hold each Contributor harmless for any liability
174
+ incurred by, or claims asserted against, such Contributor by reason
175
+ of your accepting any such warranty or additional liability.
176
+
177
+ END OF TERMS AND CONDITIONS
178
+
179
+ APPENDIX: How to apply the Apache License to your work.
180
+
181
+ To apply the Apache License to your work, attach the following
182
+ boilerplate notice, with the fields enclosed by brackets "[]"
183
+ replaced with your own identifying information. (Don't include
184
+ the brackets!) The text should be enclosed in the appropriate
185
+ comment syntax for the file format. We also recommend that a
186
+ file or class name and description of purpose be included on the
187
+ same "printed page" as the copyright notice for easier
188
+ identification within third-party archives.
189
+
190
+ Copyright [yyyy] [name of copyright owner]
191
+
192
+ Licensed under the Apache License, Version 2.0 (the "License");
193
+ you may not use this file except in compliance with the License.
194
+ You may obtain a copy of the License at
195
+
196
+ http://www.apache.org/licenses/LICENSE-2.0
197
+
198
+ Unless required by applicable law or agreed to in writing, software
199
+ distributed under the License is distributed on an "AS IS" BASIS,
200
+ WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
201
+ See the License for the specific language governing permissions and
202
+ limitations under the License.
@@ -0,0 +1,170 @@
1
+ Metadata-Version: 2.4
2
+ Name: wyoming-mlx
3
+ Version: 0.1.0
4
+ Summary: Apple-Silicon-native TTS and STT for Home Assistant and OpenAI-compatible clients
5
+ Project-URL: Homepage, https://github.com/rnorth/wyoming-mlx
6
+ Project-URL: Issues, https://github.com/rnorth/wyoming-mlx/issues
7
+ Author-email: Richard North <rich.north@gmail.com>
8
+ License-Expression: Apache-2.0
9
+ License-File: LICENSE
10
+ Keywords: apple-silicon,home-assistant,kokoro,mlx,stt,tts,whisper,wyoming
11
+ Classifier: Development Status :: 4 - Beta
12
+ Classifier: Operating System :: MacOS
13
+ Classifier: Programming Language :: Python :: 3.12
14
+ Classifier: Topic :: Multimedia :: Sound/Audio :: Speech
15
+ Requires-Python: >=3.12
16
+ Requires-Dist: fastapi
17
+ Requires-Dist: kokoro>=0.9.4
18
+ Requires-Dist: mlx-whisper>=0.4.3; sys_platform == 'darwin'
19
+ Requires-Dist: mlx>=0.31.2; sys_platform == 'darwin'
20
+ Requires-Dist: numpy
21
+ Requires-Dist: pydantic
22
+ Requires-Dist: pydantic-settings
23
+ Requires-Dist: python-multipart>=0.0.29
24
+ Requires-Dist: scipy
25
+ Requires-Dist: soundfile
26
+ Requires-Dist: torch
27
+ Requires-Dist: uvicorn[standard]
28
+ Requires-Dist: wyoming
29
+ Description-Content-Type: text/markdown
30
+
31
+ # wyoming-mlx
32
+
33
+ Apple-Silicon-native TTS (Kokoro) and STT (distil-whisper) for Home Assistant
34
+ and OpenAI-compatible clients.
35
+
36
+ ## Why?
37
+
38
+ If you have a Mac on your network, it can be your voice server. wyoming-mlx
39
+ turns it into a fast, fully local speech-to-text and text-to-speech service:
40
+
41
+ - **Private by construction.** Audio never leaves your network — no cloud
42
+ speech APIs, no per-request pricing, nothing to subscribe to. Models run
43
+ entirely on your machine.
44
+ - **Uses hardware you already own.** Apple Silicon's GPU and unified memory
45
+ run Whisper and Kokoro comfortably alongside whatever else the Mac is
46
+ doing — no dedicated GPU server, no idle power draw of a CUDA box.
47
+ - **One service, two ecosystems.** Home Assistant talks to it natively over
48
+ the [Wyoming protocol](https://github.com/rhasspy/wyoming) (drop-in
49
+ replacement for `wyoming-faster-whisper`/`wyoming-piper` satellites), while
50
+ anything that speaks the OpenAI audio API — scripts, editors, chat UIs —
51
+ can use the same instance via `/v1/audio/transcriptions` and
52
+ `/v1/audio/speech`.
53
+ - **Set-and-forget.** Install with Homebrew, run as a launchd service via
54
+ `brew services`, and models are fetched once into the Hugging Face cache.
55
+
56
+ Speech-to-text runs on [MLX](https://github.com/ml-explore/mlx) via
57
+ `mlx-whisper`; text-to-speech runs Kokoro on Metal via PyTorch. The real
58
+ backends therefore require an Apple Silicon Mac. Everything else (config,
59
+ HTTP API, Wyoming protocol handling, fake backends, tests) is portable, and
60
+ CI runs on Linux against the fake backends.
61
+
62
+ ## Install (Homebrew)
63
+
64
+ ```bash
65
+ brew tap rnorth/tap
66
+ brew install wyoming-mlx
67
+ ```
68
+
69
+ Run it in the foreground with `wyoming-mlx`, or as a launchd service:
70
+
71
+ ```bash
72
+ brew services start wyoming-mlx
73
+ ```
74
+
75
+ Logs go to `$(brew --prefix)/var/log/wyoming-mlx.log`. Apple Silicon only.
76
+
77
+ ## Quick start (dev)
78
+
79
+ ```bash
80
+ mise install
81
+ uv sync
82
+ uv run pytest
83
+ ```
84
+
85
+ Works on Linux too: CPU-only torch is selected automatically (the fake
86
+ backends need no GPU). Integration tests against real models are skipped by
87
+ default; run them with `uv run pytest --integration` (Apple Silicon only).
88
+
89
+ ## Run locally (fake backends)
90
+
91
+ ```bash
92
+ uv run python scripts/dev_run.py
93
+ ```
94
+
95
+ The dev server uses the API key `dev`.
96
+
97
+ ## Run locally (real MLX backends)
98
+
99
+ ```bash
100
+ uv run wyoming-mlx
101
+ ```
102
+
103
+ By default it loads:
104
+ - distil-whisper-large-v3 (MLX) on Wyoming port 10300 / HTTP `/v1/audio/transcriptions`
105
+ - Kokoro-82M (MLX) on Wyoming port 10200 / HTTP `/v1/audio/speech`
106
+ - HTTP on port 10400 with API-key auth
107
+
108
+ Models download on first use to the Hugging Face cache.
109
+
110
+ ### API keys
111
+
112
+ HTTP endpoints require a bearer token. Keys are read at startup from
113
+ `~/.config/wyoming-mlx/apikeys` (override with `--http-api-keys-file`),
114
+ one key per line, `#` comments allowed. The file should be mode `0600`.
115
+ If the file is missing or empty, all HTTP requests are rejected with 401.
116
+
117
+ Note that the HTTP API listens on all interfaces by default (set
118
+ `WYOMING_MLX_HTTP__HOST=127.0.0.1` to restrict it), and `GET /v1/models`
119
+ is unauthenticated, matching OpenAI API behaviour.
120
+
121
+ ```bash
122
+ mkdir -p ~/.config/wyoming-mlx
123
+ (umask 077; openssl rand -hex 32 > ~/.config/wyoming-mlx/apikeys)
124
+ ```
125
+
126
+ ## HTTP API
127
+
128
+ ### List models
129
+
130
+ ```bash
131
+ curl http://localhost:10400/v1/models
132
+ ```
133
+
134
+ ### Transcribe an audio file
135
+
136
+ ```bash
137
+ curl http://localhost:10400/v1/audio/transcriptions \
138
+ -H "Authorization: Bearer $API_KEY" \
139
+ -F file=@some.wav
140
+ ```
141
+
142
+ ### Synthesize speech
143
+
144
+ ```bash
145
+ curl http://localhost:10400/v1/audio/speech \
146
+ -H "Authorization: Bearer $API_KEY" \
147
+ -H "Content-Type: application/json" \
148
+ -d '{"input":"Hello there.","voice":"af_heart"}' \
149
+ --output /tmp/out.wav
150
+ ```
151
+
152
+ ## Home Assistant integration
153
+
154
+ Settings → Integrations → Wyoming Protocol → Add:
155
+
156
+ - STT: `<host>:10300`
157
+ - TTS: `<host>:10200`
158
+
159
+ No keys, no TLS (HA convention, trusted LAN).
160
+
161
+ ## Configuration
162
+
163
+ Pass `--config /path/to/config.toml` or set env vars with the
164
+ `WYOMING_MLX_` prefix and `__` for nesting (e.g.
165
+ `WYOMING_MLX_HTTP__PORT=10401`). See `src/wyoming_mlx/config.py` for the
166
+ full schema.
167
+
168
+ ## License
169
+
170
+ [Apache-2.0](LICENSE)
@@ -0,0 +1,140 @@
1
+ # wyoming-mlx
2
+
3
+ Apple-Silicon-native TTS (Kokoro) and STT (distil-whisper) for Home Assistant
4
+ and OpenAI-compatible clients.
5
+
6
+ ## Why?
7
+
8
+ If you have a Mac on your network, it can be your voice server. wyoming-mlx
9
+ turns it into a fast, fully local speech-to-text and text-to-speech service:
10
+
11
+ - **Private by construction.** Audio never leaves your network — no cloud
12
+ speech APIs, no per-request pricing, nothing to subscribe to. Models run
13
+ entirely on your machine.
14
+ - **Uses hardware you already own.** Apple Silicon's GPU and unified memory
15
+ run Whisper and Kokoro comfortably alongside whatever else the Mac is
16
+ doing — no dedicated GPU server, no idle power draw of a CUDA box.
17
+ - **One service, two ecosystems.** Home Assistant talks to it natively over
18
+ the [Wyoming protocol](https://github.com/rhasspy/wyoming) (drop-in
19
+ replacement for `wyoming-faster-whisper`/`wyoming-piper` satellites), while
20
+ anything that speaks the OpenAI audio API — scripts, editors, chat UIs —
21
+ can use the same instance via `/v1/audio/transcriptions` and
22
+ `/v1/audio/speech`.
23
+ - **Set-and-forget.** Install with Homebrew, run as a launchd service via
24
+ `brew services`, and models are fetched once into the Hugging Face cache.
25
+
26
+ Speech-to-text runs on [MLX](https://github.com/ml-explore/mlx) via
27
+ `mlx-whisper`; text-to-speech runs Kokoro on Metal via PyTorch. The real
28
+ backends therefore require an Apple Silicon Mac. Everything else (config,
29
+ HTTP API, Wyoming protocol handling, fake backends, tests) is portable, and
30
+ CI runs on Linux against the fake backends.
31
+
32
+ ## Install (Homebrew)
33
+
34
+ ```bash
35
+ brew tap rnorth/tap
36
+ brew install wyoming-mlx
37
+ ```
38
+
39
+ Run it in the foreground with `wyoming-mlx`, or as a launchd service:
40
+
41
+ ```bash
42
+ brew services start wyoming-mlx
43
+ ```
44
+
45
+ Logs go to `$(brew --prefix)/var/log/wyoming-mlx.log`. Apple Silicon only.
46
+
47
+ ## Quick start (dev)
48
+
49
+ ```bash
50
+ mise install
51
+ uv sync
52
+ uv run pytest
53
+ ```
54
+
55
+ Works on Linux too: CPU-only torch is selected automatically (the fake
56
+ backends need no GPU). Integration tests against real models are skipped by
57
+ default; run them with `uv run pytest --integration` (Apple Silicon only).
58
+
59
+ ## Run locally (fake backends)
60
+
61
+ ```bash
62
+ uv run python scripts/dev_run.py
63
+ ```
64
+
65
+ The dev server uses the API key `dev`.
66
+
67
+ ## Run locally (real MLX backends)
68
+
69
+ ```bash
70
+ uv run wyoming-mlx
71
+ ```
72
+
73
+ By default it loads:
74
+ - distil-whisper-large-v3 (MLX) on Wyoming port 10300 / HTTP `/v1/audio/transcriptions`
75
+ - Kokoro-82M (MLX) on Wyoming port 10200 / HTTP `/v1/audio/speech`
76
+ - HTTP on port 10400 with API-key auth
77
+
78
+ Models download on first use to the Hugging Face cache.
79
+
80
+ ### API keys
81
+
82
+ HTTP endpoints require a bearer token. Keys are read at startup from
83
+ `~/.config/wyoming-mlx/apikeys` (override with `--http-api-keys-file`),
84
+ one key per line, `#` comments allowed. The file should be mode `0600`.
85
+ If the file is missing or empty, all HTTP requests are rejected with 401.
86
+
87
+ Note that the HTTP API listens on all interfaces by default (set
88
+ `WYOMING_MLX_HTTP__HOST=127.0.0.1` to restrict it), and `GET /v1/models`
89
+ is unauthenticated, matching OpenAI API behaviour.
90
+
91
+ ```bash
92
+ mkdir -p ~/.config/wyoming-mlx
93
+ (umask 077; openssl rand -hex 32 > ~/.config/wyoming-mlx/apikeys)
94
+ ```
95
+
96
+ ## HTTP API
97
+
98
+ ### List models
99
+
100
+ ```bash
101
+ curl http://localhost:10400/v1/models
102
+ ```
103
+
104
+ ### Transcribe an audio file
105
+
106
+ ```bash
107
+ curl http://localhost:10400/v1/audio/transcriptions \
108
+ -H "Authorization: Bearer $API_KEY" \
109
+ -F file=@some.wav
110
+ ```
111
+
112
+ ### Synthesize speech
113
+
114
+ ```bash
115
+ curl http://localhost:10400/v1/audio/speech \
116
+ -H "Authorization: Bearer $API_KEY" \
117
+ -H "Content-Type: application/json" \
118
+ -d '{"input":"Hello there.","voice":"af_heart"}' \
119
+ --output /tmp/out.wav
120
+ ```
121
+
122
+ ## Home Assistant integration
123
+
124
+ Settings → Integrations → Wyoming Protocol → Add:
125
+
126
+ - STT: `<host>:10300`
127
+ - TTS: `<host>:10200`
128
+
129
+ No keys, no TLS (HA convention, trusted LAN).
130
+
131
+ ## Configuration
132
+
133
+ Pass `--config /path/to/config.toml` or set env vars with the
134
+ `WYOMING_MLX_` prefix and `__` for nesting (e.g.
135
+ `WYOMING_MLX_HTTP__PORT=10401`). See `src/wyoming_mlx/config.py` for the
136
+ full schema.
137
+
138
+ ## License
139
+
140
+ [Apache-2.0](LICENSE)
@@ -0,0 +1,3 @@
1
+ [tools]
2
+ python = "3.12"
3
+ uv = "0.10.2"