Mandarin Phonemes

The tables below list the full set of phonemes supported by Capacity Private Cloud Neural and LLM TTS for Mandarin (zh-CN). Each phoneme is shown with its Unicode code point(s), decimal value(s), IPA symbol, X-SAMPA representation, viseme, and an example word. Consonants and vowels are listed separately.

Phonemes can be supplied in SSML using the <phoneme> element (with alphabet="ipa" or alphabet="x-sampa") to override the default pronunciation of a word. For the complete list of Neural TTS voices, see Neural TTS Languages.

These phonemes apply to the following Mandarin Neural TTS voices:

Mailin
Lanying

Supplementary English phonemes. In addition to the native Mandarin inventory, this phoneme set includes a number of English phonemes (marked with ^* and grouped at the end of each table). These are provided so that English words and loanwords that appear within Mandarin text are pronounced correctly.

Consonants

Unicode	Decimal	IPA	X-SAMPA	Viseme	Example
U+0070	112	p	p	DIS_1	宝
U+02A6 U+02B0	678 688	ʦʰ	ts_h	DIS_7	策
U+0288 U+0282 U+02B0	648 642 688	ʈʂʰ	t`s`_h	DIS_8	丑
U+0074	116	t	t	DIS_7	定
U+0066	102	f	f	DIS_4	风
U+006B	107	k	k	DIS_9	购
U+0068	104	h	h	DIS_9	航
U+02A8	680	ʨ	ts\	DIS_8	结
U+006B U+02B0	107 688	kʰ	k_h	DIS_9	空
U+006C	108	l	l	DIS_6	聊
U+006D	109	m	m	DIS_1	美
U+006E	110	n	n	DIS_7	奶
U+014B	331	ŋ	N	DIS_9	旁
U+0070 U+02B0	112 688	pʰ	p_h	DIS_1	怕
U+02A8 U+02B0	680 688	ʨʰ	ts\_h	DIS_8	强
U+027B	635	ɻ	r\`	DIS_3	染
U+0073	115	s	s	DIS_7	三
U+0282	642	ʂ	s`	DIS_8	十
U+0074 U+02B0	116 688	tʰ	t_h	DIS_7	特
U+0077	119	w	w	DIS_2	外
U+0255	597	ɕ	s\	DIS_8	西
U+006A	106	j	j	DIS_10	雨
U+02A6	678	ʦ	ts	DIS_7	嘴
U+0288 U+0282	648 642	ʈʂ	t`s`	DIS_8	这
U+00F0	240	ð	D	DIS_5	then ^*
U+0072	114	r	r	DIS_3	ray ^*
U+03B8	952	θ	T	DIS_5	thin ^*
U+0076	118	v	v	DIS_4	van ^*
U+0292	658	ʒ	Z	DIS_8	beige ^*

Vowels

Unicode	Decimal	IPA	X-SAMPA	Viseme	Example
U+0061 U+0308	97 776	ä	a_"	DIS_13	怕
U+028C	652	ʌ	V	DIS_13	乐
U+025B	603	ɛ	E	DIS_11	也
U+0069	105	i	i	DIS_10	币
U+0252	594	ɒ	Q	DIS_13	伯
U+0075	117	u	u	DIS_3	路
U+0079	121	y	y	DIS_3	据
U+0264	612	ɤ	7	DIS_9	之
U+0061 U+0308 U+0069	97 776 105	äi	a_"i	DIS_13	买
U+0061 U+0308 U+0075	97 776 117	äu	a_"u	DIS_13	老
U+0065 U+0069	101 105	ei	ei	DIS_11	肥
U+028C U+02DE	652 734	ʌ˞	V`	DIS_13	二
U+0069 U+0061 U+0308	105 97 776	iä	ia_"	DIS_10	下
U+0069 U+025B	105 603	iɛ	iE	DIS_10	姐
U+0069 U+006F U+031E	105 111 798	io̞	io_o	DIS_10	熊
U+0069 U+0075	105 117	iu	iu	DIS_10	流
U+006F U+031E U+0075	111 798 117	o̞u	o_ou	DIS_13	后
U+0075 U+0061 U+0308	117 97 776	uä	ua_"	DIS_3	抓
U+0079 U+026A	121 618	yɪ	yI	DIS_13	虐
U+0075 U+0069	117 105	ui	ui	DIS_3	醉
U+0075 U+0252	117 594	uɒ	uQ	DIS_3	弱
U+0251	593	ɑ	A	DIS_13	palm ^*
U+025B U+0259	603 601	ɛə	E@	DIS_11	square ^*
U+026A	618	ɪ	I	DIS_10	kit ^*
U+026A U+0259	618 601	ɪə	I@	DIS_10	near ^*
U+0254 U+026A	596 618	ɔɪ	OI	DIS_12	choice ^*
U+0254	596	ɔ	O	DIS_13	thought ^*
U+028A	650	ʊ	U	DIS_11	foot ^*
U+028A U+0259	650 601	ʊə	U@	DIS_11	cure ^*

^* Supplementary English phoneme — included so that English words and loanwords embedded in Mandarin text render with correct pronunciation.

Viseme values use the standard Capacity viseme set (for lip-sync and avatar animation). X-SAMPA is the ASCII-encoded equivalent of the IPA symbol.

Was this article helpful?