bluejay/bluejay-infra
1
0
Fork 0
Code Issues Pull Requests 14 Actions Packages Projects Releases Wiki Activity

fc-ttsreader: enable real Whisper alignment via fc-speech-align

Browse Source 
Flips Speech__Alignment__Enabled=true and points BaseUrl at the
new BLUEJAY-WS podman quadlet running fc-speech-align (faster-
whisper, /align contract). When Lane 1δ's
/api/v1/voices/preview-with-timings runs after this lands, the
alignment.source field flips from 'estimated' to 'whisper' and
the per-word timings come from real audio analysis instead of
uniform-spacing estimates.

No image rebuild — the Lane 1α DI registration already routes
IWhisperAlignmentClient to FasterWhisperAlignmentClient when
Speech:Alignment:Enabled is true.

Companion firewall rule from FlowerCore.Puppet@bbc02ea +
@05504ed (whisper_align_enabled flag on bluejay-ws-linux Hiera)
opens port 9200 to RKE2 pod CIDR durably.
This commit is contained in:
Andrew Stoltz
2026-04-24 23:37:30 -05:00
parent a760a58846
commit 76ece92cfd
1 changed files with 6 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
apps/fc-ttsreader/fc-ttsreader.yaml
View File

@@ -172,6 +172,12 @@ spec:
value: "http://10.0.56.20:10401" value: "http://10.0.56.20:10401"
- name: TtsReader__Kokoro__TimeoutSeconds - name: TtsReader__Kokoro__TimeoutSeconds
value: "120" value: "120"
- name: Speech__Alignment__Enabled
value: "true"
- name: Speech__Alignment__BaseUrl
value: "http://10.0.56.20:9200"
- name: Speech__Alignment__TimeoutSeconds
value: "120"
- name: TtsReader__Ollama__BaseUrl - name: TtsReader__Ollama__BaseUrl
value: "http://10.0.57.17:11434" value: "http://10.0.57.17:11434"
- name: TtsReader__Ollama__DefaultModel - name: TtsReader__Ollama__DefaultModel