doubao-tts

Name: doubao-tts
Author: calesthio

295installs4.9kforks

How to install doubao-tts?

calesthio/openmontage/doubao-tts

$npx -y skills add calesthio/OpenMontage --skill doubao-tts

Installs into the current project.

Summary

Generate Mandarin and multilingual narration with Volcengine Doubao Speech 2.0. Use when creating Chinese voiceovers, when the user prefers Doubao/Volcengine/火山引擎/豆包 TTS, or when narration needs character-level timestamp metadata for subtitles.

Files · 1

View on GitHub

SKILL.md

1	# Doubao TTS
2
3	Requires `DOUBAO_SPEECH_API_KEY` in `.env`.
4	Set `DOUBAO_SPEECH_VOICE_TYPE` for the default voice, or pass `voice_id` to the tool.
5
6	## Current API
7
8	Use the new-console API key flow:
9
10	```text
11	X-Api-Key: ${DOUBAO_SPEECH_API_KEY}
12	X-Api-Resource-Id: seed-tts-2.0
13	```
14
15	Do not use `X-Api-App-Id` and `X-Api-Access-Key` with a new-console API Key. If the API returns `load grant: requested grant not found`, the key type or auth header is probably wrong.
16
17	For long-form video narration, prefer the async endpoint:
18
19	```text
20	POST https://openspeech.bytedance.com/api/v3/tts/submit
21	POST https://openspeech.bytedance.com/api/v3/tts/query
22	```
23
24	This returns `audio_url` plus `sentences[].words[]` timing metadata that can be used to build subtitles.
25
26	## OpenMontage Usage
27
28	Generate with the TTS selector:
29
30	```python
31	from tools.audio.tts_selector import TTSSelector
32
33	result = TTSSelector().execute({
34	"preferred_provider": "doubao",
35	"text": "如果 AI 真的会改变未来，普通人到底该怎么参与？",
36	"voice_id": "zh_female_vv_uranus_bigtts",
37	"output_path": "projects/my-video/assets/audio/narration.mp3",
38	"speech_rate": 0,
39	"enable_timestamp": True,
40	})
41	```
42
43	Or call the provider directly:
44
45	```python
46	from tools.audio.doubao_tts import DoubaoTTS
47
48	result = DoubaoTTS().execute({
49	"text": "短样本试听文本。",
50	"voice_id": "zh_female_vv_uranus_bigtts",
51	"output_path": "projects/my-video/assets/audio/doubao_sample.mp3",
52	})
53	```
54
55	The provider writes:
56
57	- `output_path`: downloaded audio file
58	- `metadata_path`: full query response JSON, defaulting to `<output_path>.json`
59
60	## Recommended Workflow
61
62	1. Generate a 10-15 second sample before a full paid narration.
63	2. Ask the user to approve voice naturalness, accent, and speed.
64	3. Generate the full narration only after approval.
65	4. Keep the query JSON. It is the source of truth for subtitle timing.
66	5. Build captions from `sentences[].words[]`, not from estimated text length.
67	6. Group captions by Chinese semantic phrases before applying timestamps. Do not split only by fixed character count; it can break phrases like "在不押单个公司的情况下" or "可能会被慢慢稀释" and hurt comprehension.
68	7. Let the video duration follow the approved voice rhythm unless the user explicitly asks to match a prior runtime.
69
70	## Parameters
71
72	- `voice_id`: Doubao `speaker` / voice type. Defaults to `DOUBAO_SPEECH_VOICE_TYPE`.
73	- `resource_id`: use `seed-tts-2.0` for Doubao Speech 2.0 voices.
74	- `speech_rate`: `0` is normal, `100` is 2x, `-50` is 0.5x.
75	- `sample_rate`: default `24000`.
76	- `enable_timestamp`: default `true`.
77	- `return_usage`: default `true`, requests usage metadata when available.
78
79	Do not pass `additions.explicit_language` by default. Some endpoint/key combinations reject `zh-cn` with `unsupported additions explicit language zh-cn`.
80
81	For calm Mandarin explainers, start with `speech_rate: 0`. If the result is too long for the approved format, make a short comparison sample with `speech_rate: 25` or `50` before regenerating the full narration. Do not speed up only to match a previous provider's duration if the user prefers Doubao's natural pace.
82
83	## Troubleshooting
84
85	- `load grant: requested grant not found`: wrong key type or wrong auth header. Use `X-Api-Key` for new-console API Keys.
86	- `speaker permission denied`: voice id is wrong or not authorized for the selected resource.
87	- `quota exceeded`: quota, lifetime characters, or concurrency exceeded.
88	- Missing timestamps: verify `enable_timestamp: true`, keep the query JSON, and confirm the selected endpoint returned `sentences`.
89
90	## Safety
91
92	Never print or write the API key to logs, metadata, patches, or project artifacts. `.env.example` should contain only empty variable names.

People who liked this also viewed

6 picks

Type

doubao-tts

295installs4.9kforks

View on GitHub

How to install doubao-tts?

calesthio/openmontage/doubao-tts

$npx -y skills add calesthio/OpenMontage --skill doubao-tts

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# Doubao TTS
2
3	Requires `DOUBAO_SPEECH_API_KEY` in `.env`.
4	Set `DOUBAO_SPEECH_VOICE_TYPE` for the default voice, or pass `voice_id` to the tool.
5
6	## Current API
7
8	Use the new-console API key flow:
9
10	```text
11	X-Api-Key: ${DOUBAO_SPEECH_API_KEY}
12	X-Api-Resource-Id: seed-tts-2.0
13	```
14
15	Do not use `X-Api-App-Id` and `X-Api-Access-Key` with a new-console API Key. If the API returns `load grant: requested grant not found`, the key type or auth header is probably wrong.
16
17	For long-form video narration, prefer the async endpoint:
18
19	```text
20	POST https://openspeech.bytedance.com/api/v3/tts/submit
21	POST https://openspeech.bytedance.com/api/v3/tts/query
22	```
23
24	This returns `audio_url` plus `sentences[].words[]` timing metadata that can be used to build subtitles.
25
26	## OpenMontage Usage
27
28	Generate with the TTS selector:
29
30	```python
31	from tools.audio.tts_selector import TTSSelector
32
33	result = TTSSelector().execute({
34	"preferred_provider": "doubao",
35	"text": "如果 AI 真的会改变未来，普通人到底该怎么参与？",
36	"voice_id": "zh_female_vv_uranus_bigtts",
37	"output_path": "projects/my-video/assets/audio/narration.mp3",
38	"speech_rate": 0,
39	"enable_timestamp": True,
40	})
41	```
42
43	Or call the provider directly:
44
45	```python
46	from tools.audio.doubao_tts import DoubaoTTS
47
48	result = DoubaoTTS().execute({
49	"text": "短样本试听文本。",
50	"voice_id": "zh_female_vv_uranus_bigtts",
51	"output_path": "projects/my-video/assets/audio/doubao_sample.mp3",
52	})
53	```
54
55	The provider writes:
56
57	- `output_path`: downloaded audio file
58	- `metadata_path`: full query response JSON, defaulting to `<output_path>.json`
59
60	## Recommended Workflow
61
62	1. Generate a 10-15 second sample before a full paid narration.
63	2. Ask the user to approve voice naturalness, accent, and speed.
64	3. Generate the full narration only after approval.
65	4. Keep the query JSON. It is the source of truth for subtitle timing.
66	5. Build captions from `sentences[].words[]`, not from estimated text length.
67	6. Group captions by Chinese semantic phrases before applying timestamps. Do not split only by fixed character count; it can break phrases like "在不押单个公司的情况下" or "可能会被慢慢稀释" and hurt comprehension.
68	7. Let the video duration follow the approved voice rhythm unless the user explicitly asks to match a prior runtime.
69
70	## Parameters
71
72	- `voice_id`: Doubao `speaker` / voice type. Defaults to `DOUBAO_SPEECH_VOICE_TYPE`.
73	- `resource_id`: use `seed-tts-2.0` for Doubao Speech 2.0 voices.
74	- `speech_rate`: `0` is normal, `100` is 2x, `-50` is 0.5x.
75	- `sample_rate`: default `24000`.
76	- `enable_timestamp`: default `true`.
77	- `return_usage`: default `true`, requests usage metadata when available.
78
79	Do not pass `additions.explicit_language` by default. Some endpoint/key combinations reject `zh-cn` with `unsupported additions explicit language zh-cn`.
80
81	For calm Mandarin explainers, start with `speech_rate: 0`. If the result is too long for the approved format, make a short comparison sample with `speech_rate: 25` or `50` before regenerating the full narration. Do not speed up only to match a previous provider's duration if the user prefers Doubao's natural pace.
82
83	## Troubleshooting
84
85	- `load grant: requested grant not found`: wrong key type or wrong auth header. Use `X-Api-Key` for new-console API Keys.
86	- `speaker permission denied`: voice id is wrong or not authorized for the selected resource.
87	- `quota exceeded`: quota, lifetime characters, or concurrency exceeded.
88	- Missing timestamps: verify `enable_timestamp: true`, keep the query JSON, and confirm the selected endpoint returned `sentences`.
89
90	## Safety
91
92	Never print or write the API key to logs, metadata, patches, or project artifacts. `.env.example` should contain only empty variable names.

People who liked this also viewed

6 picks

Type